Best distortion pedals for bass 2022: thicken up your low end with these aggressive bass stompboxes Guitar World
source
A pretraining domain decomposition method using artificial neural networks to solve elliptic PDE boundary value problems | Scientific Reports – Nature.com
Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.
Advertisement
Scientific Reports volume 12, Article number: 13939 (2022)
Metrics details
Developing methods of domain decomposition (DDM) has been widely studied in the field of numerical computation to estimate solutions of partial differential equations (PDEs). Several case studies have also reported that it is feasible to use the domain decomposition approach for the application of artificial neural networks (ANNs) to solve PDEs. In this study, we devised a pretraining scheme called smoothing with a basis reconstruction process on the structure of ANNs and then implemented the classic concept of DDM. The pretraining process that is engaged at the beginning of the training epochs can make the approximation basis become well-posed on the domain so that the quality of the estimated solution is enhanced. We report that such a well-organized pretraining scheme may affect any NN-based PDE solvers as we can speed up the approximation, improve the solution’s smoothness, and so on. Numerical experiments were performed to verify the effectiveness of the proposed DDM method on ANN for estimating solutions of PDEs. Results revealed that this method could be used as a tool for tasks in general machine learning.
Partial differential equations (PDEs) contain principled physical laws and govern the object’s physical or engineering systems. For instance, PDEs are mathematically analyzed to understand scientific and physical phenomena such as heat, electricity, fluid dynamics, and so on. However, due to the analytic complexity of a given PDE, it is usually impossible to have their explicit solutions. In correspondence, in the science of numerical solutions of PDEs, many studies have been conducted to efficiently solve problems by applying various discretization methods such as the finite difference method1, finite element method2, and other high-order methods such as the spectral method3. Eventually, as the era of the artificial intelligence approaches4 with their applications, for example, to cognitive science5, biomedical science6,7, ecology8,9, manufacturing10, and more, many numerical methodologies of computational physics based on machine learning techniques have also been actively studied, including radial basis function networks11, Gaussian Kernels12, least squares support vector machine13, dimension reduction methods14,15,16, deep learning with Monte Carlo method17, autoencoder18, and convolutional neural networks19.
On the other hand, Li et al.20 have briefly summarized the use of machine learning into three categories: (1) implementing physics into machine learning by generating a dataset of physical laws21, (2) designing a physics-guided machine learning method with physics-guided model as a black box22, and (3) imposing physics into the training process, such as the method of “physics-informed neural networks (PINN)”23. Among these various types of machine learning technologies, this paper will focus on artificial neural networks (ANN) of the third category24,25,26. Unlike most of other state-of-the-art machine learning techniques that lack robustness or fail to guarantee convergence with small data regimes, the method of PINN encodes physical information into the training process and well generalizes to the right solution even when only a few training examples are available23.
Regarding the development of ANN-based PDE solvers, Lagaris et al.27,28 have made early contributions to the application of ANN towards solving systems of PDEs. Using analogues of their model, subsequent research has been performed on empirical methodologies to use ANNs for the calculus of PDEs, including irregular boundary problems, arbitrary boundary problems29,30, initial-boundary value problems31, time-constrained optimal control problems32, the methods of unsupervised learning33, immersed boundary problem34, and a constrained backpropagation approach35.
In 2020, devising a domain decomposition methodology (DDM), Jagtap et al.36 proposed conservative physics-informed neural networks to solve Burgers’ equations, incompressible Navier–Stokes equations, and compressible Euler equations. Among many variations of ANN models, in Refs.36,37, the authors implemented PINN-based domain decomposition methodologies and showed their approximations’ effectiveness by hp-refinement. They divided their computational domain into several subdomains and constructed the main loss function by employing weight parameters on terms of boundary/initial conditions, the residual from governing equations and interfaces, and applied their method to forward and inverse problems of several challenging nonlinear PDEs. Since hp-refinement technologies have been one of the main concerns for research performing numerical analysis of PDEs38,39,40, their breakthroughs employing the ANN-based decomposition of the domain could not be regarded as anything other than impressive.
Solving PDE by ANN models, many studies have developed efficient technologies as listed in above. However, there is no model that employs a concept of pretraining technique. As the main contribution of our proposed method in our work, we proposed a pretraining scheme to improve the performance of a given model of ANN to solve elliptic PDEs.
In this study, as another approach for an ANN-based DDM, we developed a pretraining scheme DDM. The pretraining process that is engaged in at the beginning of the training epochs helps the approximation basis get well-posed on the domain so that the accuracy of the estimated solution is enhanced, i.e., the speed of approximation is improved rather than non-pretrained schemes. We report that such a well-organized pretraining scheme is effective as an NN-based PDE solver. In our work, we did not focus on handling disadvantages of existing methods. Instead, we focused on developing an ad-hoc technology by proposing a pretraining scheme so that it could be applied to any given existing ANN models to enhance their numerical performances.
The proposed pretraining scheme consists of two parts. The first part is to train along with governing equations without any engagement of the boundary or subdomain’s interfacial conditions. The second part is to train for neighboring subdomains to be flattened, i.e., a subdomain’s function of the estimation is compelled to be zero at data points located on neighboring subdomains. In this article, we call the first part the smoothing process and the second part the basis reconstruction (BR) process. With numerical experiments, we show that as the proposed pretraining process continues, the estimated solution’s output of the fine-tuning improves. Moreover, we report that the BR process helps the numerical basis of the ANN become well-posed to approximate sorts of high degrees of polynomials that are in solutions as the BR process makes the basis of ANN of a subdomain closely gather on its region such that the initial state of the ANN, at the starting point of the fine-tuning, has high smoothness.
Other than existing methodologies of the NN-based domain decomposition method that employ hyperparameters within the loss function, our proposed method consists of a scheme of pretraining and fine-tuning approach that allows us to use the scheme in various architectures of deep-learning algorithms. As the main contribution of our proposed method, the pretraining process that is engaged at the beginning of the training epochs helps the approximation basis being well-posed on the domain so that the quality of the estimated solution is enhanced. We report that such well-organized pretraining scheme may give effectiveness to any NN-based PDE solver as we get a speed-up in approximation and improvement in the solution’s smoothness and so on.
In “Methods” section, we introduce our domain decomposition method for ANNs. In “Numerical experiments” section, we present numerical results of using our method to solve elliptic boundary value PDEs and the XOR problem as a demonstration of the method’s performance for application to the general machine learning task. We then discuss properties of our proposed method. In “Conclusion” section, we present some conclusions drawn from our study.
Let the architecture of the multi-layered ANN (abbreviated as ANN thereafter) be given as follows:
where (vin {mathbb{R}}^{H}) and (win {mathbb{R}}^{Htimes left(d+1right)}) denote learnable weights, (p) represents them as general parameters, the input variable is denoted by (overline{x }), an extension of (x) such that (overline{x }={left(x,bright)}^{T}) with a constant (b) for the computation of the bias weight in (w), and “(sigma :{mathbb{R}}^{H}to {mathbb{R}}^{H})” is a component-wise activation function. In our study, we employed the sigmoidal activation function. The (d)-dimensional problem structured by (Nleft(x,pright)) is illustrated in Fig. 1.
(d)-Dimensional ANN structure.
Following28, we considered the following system of PDEs:
where (Psi left(xright)), with (x=left(x(1),cdots ,x(d)right)in {mathbb{R}}^{d}), was the solution of Eq. (2) on a domain (Omega subset {mathbb{R}}^{d}) which satisfied essential boundary conditions specified on the boundary (partialOmega) of (Omega). In our numerical examples, we supposed (Omega) to be an open convex polygon in ({mathbb{R}}^{2}) such that (overline{Omega }) was the union of a finite number of polyhedra.
To satisfy Eq. (2) and approximate (Psi left(xright)), let us define a trial solution ({Psi }_{t}left(xright):)
where (Aleft(xright)) satisfies the boundary condition such that (Psi left(xright)=Aleft(xright), forall xin partialOmega), and (Nleft(x,pright)) is the single-hidden-layered ANN. The scalar function (Bleft(xright)) is constructed to satisfy (Bleft(xright)Nleft(x,pright)=0, forall xin {partialOmega }_{D}), where ({partialOmega }_{D}) is the portion of (partialOmega) where the essential boundary condition is imposed. For details of techniques on how to impose constraints, please refer to Refs.28,29.
To estimate the trial solution ({Psi }_{t}left(x,pright)) solving the system of Eq. (2), we defined discretization (widehat{Omega }) of (overline{Omega }) such that (widehat{Omega }=left{{xi }_{j}inOmega ;j=1,dots ,mright}). Then Eq. (2) is in the form of the collocation method3 as follows:
To train weight parameters of the minimization problem in Eq. (4), we applied the backpropagation method used in gradient descent27,28,35.
To compute the gradient of ({Psi }_{text{t}}left(xi ,pright)), (xi in widehat{Omega }), we discretized the 1st- and 2nd-order derivatives by the backward and midpoint Euler scheme, respectively, such that
where for a positive real number (delta >0), ({xi }_{i,-delta }) and ({xi }_{i,+delta }) were defined as varying the (i)th component of (xi) by (pm delta), i.e., ({xi }^{i,-delta }=(xi (1),xi (2),dots ,xi (i)-delta ,dots ,xi (d))), ({xi }^{i,+delta }=(xi left(1right),xi left(2right),dots ,xi left(iright)+delta ,dots ,xi (d))). Note that, for the implementation of numerical experiments in this paper, we set (delta ={10}^{-6}). Using the numerical approximation theory of derivatives, one may set it as sufficiently small.
As for our example problem of Eq. (2), the Poisson equation is given as follows:
where (Fleft(xright)) is the source function.
To estimate the solution of Eq. (6), we discretized the minimization problem of Eq. (6) as follows:
Now, by employing the gradient descent method with a suitable learning rate, we obtained the iterative solution of Eq. (7). For weight parameters (w) and (v) of (p), using Eqs. (5) and (7), their (n) th epoch’s updates of gradients (Delta {w}^{(n)}) and (Delta {v}^{(n)}) are given by the chain rule of the derivatives of Eq. (7) for ({w}^{(n)}) and ({v}^{(n)}), respectively, as follows:
where ({sigma }_{k}^{^{prime}}) is the derivative of the (k) th component of (sigma).
In our study, the given computational domain was divided into several non-overlapping smaller computational regions over which the governing PDEs were inherited. We call these small-sized computational domains as subdomains of our domain decomposition method.
Now, the core of the DDM scheme we proposed began with the flux-continuity on the interfaces of the subdomains. After that, the construction of the basis related to the solutions on the subdomains was considered. To illustrate the scheme, we defined a pair of consecutive neighborhood subdomains (widehat{{Omega }_{r}}), (widehat{{Omega }_{s}}subset widehat{Omega }) where (widehat{{Omega }_{r}}) and (widehat{{Omega }_{s}}) were disjoint except for their interfaces. Trial functions ({Psi }_{text{t},text{r}}left(cdot ,{p}_{r}right)) and ({Psi }_{text{t},text{s}}left(cdot ,{p}_{s}right)) were given for (widehat{{Omega }_{r}}) and (widehat{{Omega }_{s}}), respectively.
The training for the interfacial flux-continuity was divided into two parts: the nodal continuity and the continuity of the normal directional derivatives.
Nodal continuity:
Continuity of the outward normal directional derivatives:
For normal directional derivatives in Eq. (10), we applied direct differentiation on ({Psi }_{text{t},text{r}}left(zeta ,{p}_{r}right)) and ({Psi }_{text{t},text{s}}left(zeta ,{p}_{s}right)) for respect to normal directions on their domains. Neglecting the subscript of (r) or (s), we have
Hence,
where (overrightarrow{n}) is the outward directional normal vector on the interface.
As the key subject of our study, a pre-processing approach was taken for the model of multi-subdomains by proposing a pretraining scheme that could initialize training weight parameters well. We aimed to enhance the convergence power. This pretraining can be divided into two parts:
Pretraining for the loss function on each subdomain (smoothing process) In this step, we strengthened the convergence power interior of each subdomain except for terms of the flux-continuity. From this, we performed an early estimation of the smoothness interior before the interfacial flux-continuity exchanges. The pretraining scheme implemented to optimize the problem is given as follows:
where the subscript (q) denotes the arbitrary index of subdomains.
Pretraining to impel the basis into each subdomain (basis reconstruction (BR) process) In this step, we aim to train the network to nullify the basis arising in the neighboring subdomains. We hypothesize that if the function-shaping basis exists outside of the interface, it may weaken the power of the polynomial basis that shapes the interior estimations. Hence, to prevent the basis from being non-zero outside of the interfaces, we performed nullification pretraining such that
where the (widehat{{Omega }_{q}^{upepsilon }}) was the (upepsilon)-expansion of (widehat{{Omega }_{q}}) for a given positive number (epsilon >0in {mathbb{R}}) defined by
The definition of (widehat{{Omega }_{q}^{upepsilon }}) is illustrated in Fig. 2. Assuming that a two-dimensional (2D) rectangular domain (widehat{Omega }) is given, the gray-shaded region depicts (widehat{{Omega }_{q}}) and the region enclosed by a red curve represents (widehat{{Omega }_{q}^{upepsilon }}).
A 2D rectangular domain (widehat{Omega }) and (widehat{{Omega }_{q}^{upepsilon }}).
Note that the above two parts of the pretraining occurred subsequently in each epoch. The main purpose of the pretraining scheme is to enhance computational smoothness of basis functions made of activation functions. As we nullified solutions outside of a given subdomain, we assumed that the smooth region in the basis function moved near the region of the subdomain so that the well-posed basis into the domain might prevent leakage of the computational basis’ smoothness off the domain.
The proposed ANN version of DDM is illustrated in Fig. 3.
The structure of the pretraining version of ANN-based DDM.
The first numerical experiment we performed was a preliminary study to investigate the pretraining’s effect on one-dimensional (1D) problems. We applied the smoothing process as pretraining without basis reconstruction or nullification for the Poisson problem with the Dirichlet boundary condition.
Note that in our work, the benchmark algorithm, which is conceptually the same as the one with the algorithm introduced in Ref.36, i.e., a DDM model without pretraining process, was one of the recently proposed algorithms to apply the traditional DDM method calculated by preserving the flux continuity between subdomains to ANN as it is (refer to refs36,37.).
The 1D Poisson equation with Dirichlet boundary condition is:
where the solution (u) satisfies equations over the domain (Omega =left[text{0,2}right]) and the boundary (partialOmega) of (Omega) is (partialOmega =Omega backslash (text{0,2})). As a test solution, we let (u=uleft(xright)), for (xinOmega), such as
Figure 4 shows plots of the estimated solution made from the ordinary single domain method. They were compared to the exact solution. To compare, we verified the use of hidden units which we set to be 150, 200, and 300. To organize the set of training data, we discretized the domain equi-spatially into (3times {2}^{4}) nodes. Among estimations, only the case of using 300 hidden units gave a good approximation to the solution. Note that when updating neural weights in the training, we conducted the smallest batch model that updated weights for each input data, like the stochastic batch mode.
Solutions with 150, 200, and 300 hidden units from the 1D domain with (3times {2}^{4}) nodes as estimated by the single domain method.
Now, to assess the computing power effectiveness of the DDM model, we conducted training with a three-subdomain method without any pretraining process. Figure 5 shows a plot of these results. Here, hidden units are set to be 20, 30, and 100 on each subdomain. Hence, the total number of the hidden units is 150. As show in Fig. 5, preset hidden units of the DDM failed to estimate the solution. Thus, the ordinary scheme of the DDM with ANN did not improve the results further in this case.
The solution without pretraining process on the 1D domain is decomposed into three subdomains with ({2}^{4}) nodes in each subdomain, where the total number of hidden units is 150.
Next, to improve DDMs by using smoothing for pretraining, we applied it to the same structure of subdomains with the same sets of hidden units. The pretraining used 10,000 epochs. For fine-tuning, the ratio of the training epochs at the boundary, subdomain’s interface, and the governing equation was set to be 20:2:1, i.e., the training took place at boundary nodes for 20 epochs, then for 2 epochs at the interface, and then 1 epoch for the governing equation on interior nodes of each subdomain. Here, the number of training epochs at the boundary was set relatively high since we noted that the structure of our ANN was not equipped with an ansatz employed to hold the boundary condition. We propose that further training at the boundary will allow estimations to be stable and maintain their analytic uniqueness. The resulting configuration of the estimated solution and the plots of the progression of the loss versus epoch are shown in Figs. 6 and 7, respectively. It was revealed that the model of smoothing well approximated the exact solution in Eq. (17). However, the single domain method and the ordinary approach of the DDM failed to make an accurate estimate (Figs. 6, 7). This demonstrates that such highly-oscillatory solutions in Eq. (17), which the ordinary single domain method hardly solves with lower amounts of hidden units, are more effectively solved by the DDM model with its effective alignment of the subdomains’ numerical basis delivered from the DDM’s inherent domain separation scheme property.
The solution with the smoothing pretraining process on the 1D domain is decomposed into three subdomains with ({2}^{4}) nodes for each subdomain, where the total number of hidden units is 150.
A comparison of the decay of the cost function’s loss for the training of the solution on the 1D domain decomposed into three subdomains with ({2}^{4}) nodes for each subdomain, where the total number of hidden units is 150.
Note that, for our numerical results, the learning rate is set to be ({10}^{-6}). From our preliminary tests, the convergence rate deteriorated for some scales of the learning rates much greater than ({10}^{-6}). Even for DDM solutions, in the subregion where oscillation or size change of the solution was large, the convergence rate was significantly lower than that in other regions.
In the 1D problem, we have defined the area of the domain to reach a relatively longer region, i.e., we have defined the domain as [0, 2]. In this case of domain region, we found that the ordinary ANN with some small numbers of hidden units failed to converge to the right solution. For this reason, as a special case of the solution to reveal the effectiveness of DDM applied for the solution type made of ANN structure, ANN solution made of some hundreds of hidden units was used for comparison to show the effectiveness of DDM. In the 1D problem, we found that, even for a hardly learnable solution due to some touch conditions regarding the domain, DDM could technically provide guarantees to reach the right convergence.
In this section, numerical results for 2D elliptic PDEs are presented.
As a test problem, we solved the type of Poisson equations given in Eq. (16) with Dirichlet boundary condition for the domain (Omega ={left[text{0,1}right]}^{2}). As our test solution, we defined (u=uleft(xright)), for (x=({x}_{1},{x}_{2})inOmega), as follows:
For the structure of the ANN on each subdomain, we set the number of hidden units to be 6, 8, or 10 and applied drop-out41 of 1, 3, or 5, respectively. As in the case of the 1D problem, we separated the domain and employed the same amount of hidden units and drop-out on subdomains for each test. The parameter (upepsilon) of the BR’s nullifying range was set to be 0.05, 0.1, 0.2, 0.3, and 0.4. Settings of the ANN on each subdomain for the experiments are listed in Table 1. Note that the initial weight parameters of the ANN were given randomly since we intended to verify prominent and consistent effects of methods even in random situations. The ratio of the training epochs at the boundary, subdomain’s interface, and the governing equation was set to be 40:2:2, i.e., the training took place at boundary nodes for 40 epochs, then for 2 epochs at the interface, and then 2 epochs for the governing equation on the interior nodes of each subdomain.
The basis reconstruction scheme on the (2times 2) square subdomains of our test problem is illustrated in Fig. 8.
The basis reconstruction scheme with parameter (upepsilon) on a (2times 2) square subdomain structure.
In Fig. 9, the estimated solutions from the case of 6 hidden units are depicted in the subplots (on the left-hand side) with a decaying process of the loss function (on the right-hand side). The solution estimated on the single domain model with 6 hidden units is denoted as ‘1-domain’. The ordinary DDM model without pretraining is ‘No-pretraining’. The pretraining model of DDM with only smoothing is ‘Smoothing’. DDMs of BR methods with (upepsilon) of 0.05, 0.1, 0.2 are indicated as ‘BR-0.05’, ‘BR-0.1’, and ‘BR-0.2’, respectively. For the case of ‘1-domain’, the solution estimation was incomplete and the loss remained high, suggesting that shortages of the numerical basis in the ANN caused the estimation to fail. For other methods using DDM structures, the loss declined. These methods relaxed the shortages in the approximation basis of the single domain model. Hence, from a computational effectiveness perspective, when one applies parallel processors, DDM models might improve the convergence rate as independent calculators (processors) are distributed to subdomains, which may save training costs per processor as each processor’s allocated portion of the training data is reduced as the domain is divided (i.e., the overall order of the basis is strengthened whereas the training load on each processor is relieved)39,40.
Results of using six hidden units (left) and a comparison of the decay of the loss (right) for both the 1-domain and 4-subdomain methods (no-pretraining, smoothing, BR-0.05, BR-0.1, BR-0.2).
To demonstrate further, in Figs. 10 and 11, configurations of estimated solutions and the loss along epochs are given for cases with 8 and 10 hidden units, respectively. It was found that the single domain approximation method still struggled to reduce the loss. DDM methods were comparatively more computationally powerful, although the dominance among them was indiscriminate when the hidden layer’s size was increased.
Results for 8 hidden units (left) and comparison of the decay of loss (right) for both the 1-domain and 4-subdomain methods (no-pretraining, smoothing, BR-0.05, BR-0.1, BR-0.2).
Results for 10 hidden units (left) and comparison of the decay of loss (right) for both the 1-domain and 4-subdomain methods (no-pretraining, smoothing, BR-0.05, BR-0.1, BR-0.2).
As shown in Figs. 9, 10 and 11, the DDM-based lower-order ANN model had a more convergent solution than a higher-order single domain model. When training is performed using a given input data, computational costs for both the DDM-based model and the single domain model would be the same if both models’ numbers of hidden nodes are the same. In our work, we focused on the computational efficiency of our proposed DDM technique since single-domain models might have difficulty to obtain the right solutions for some problems while our proposed methods were validated to handle these problems.
Results of comparing different DDM models are presented in Fig. 12, which shows resulting values of the loss at the end of each training with increasing number of hidden units. As shown in Fig. 12, when the size of hidden layers is increased, results are improved. However, the problem is that weight parameters that generate hidden units’ values might be poorly posed with only a few hidden layers. This causes the training process to have shortages of numerical bases. Especially for the BR method with (upepsilon) of 0.3 and 0.4, values of the loss were highly unstable. This suggests that the value of (upepsilon) may affect the ANN’s ability to estimate accurately. Of course, more reliable evidence of (upepsilon)’s effect needs to be established by performing relevant well-developed experiments. In the next subsection, we will present that, by pretraining with the BR method, the smoothness of the basis function generated by the hidden layer is improved. Hence, we can interpret that the effect of (upepsilon) can cause the BR method to enrich the order of polynomials which shape the basis.
Comparison of the 1-domain method and the 4-subdomain method (no-pretraining, smoothing, BR-0.05, BR-0.1, BR-0.2, BR-0.3, BR-0.4).
We confirmed effects of using the ({H}_{2}) semi-norm to compare degrees of polynomials in the basis as a measure of smoothness according to pretraining. Although we did not investigate all of the basis-degree with the measure in the ({H}_{2}) semi-norm reflecting indirect comparison of the strength of the basis, our numerical results gave us confidence that the proposed pretraining method might help the initial basis reform its polynomials to higher degrees.
From the structure of the ANN, we know that the solution is a linear combination of the outputs of activated hidden units. Hence, the ({H}_{2}) semi-norm of functions shaped by the activation of the hidden units, i.e., the measure of the degree of the polynomials of each hidden unit’s activation function, can be considered as a measure of the degree of the polynomials of the numerical basis of the ANN. Precisely, we measured the ({H}_{2}) semi-norm of one hidden unit’s output values activated in each subdomain. The sum of the values of the hidden units on their subdomains was then added to make a total measure (see Table 2). In Table 2, the mean of the ({H}_{2}) semi-norm is listed as the larger number of hidden units used in the BR method will naturally increase values of the ({H}_{2}) semi-norm (H) compared to the smoothing method. In Figs. 13 and 14, values listed in Table 2 are plotted. Figure 13 (where the x-axis is H) clearly shows that the values of the ({H}_{2}) semi-norm for the BR method are almost more than those for the smoothing method. In Fig. 14, the x-axis denotes the method of pretraining. Values tended to increase as (upepsilon) decreased. This means that the basis of BR with low values of (upepsilon) has improved its polynomial degrees as a result of the pretraining. Results shown in Fig. 12 (where the BR method with (upepsilon =0.3) and 0.4 generated a relatively low-quality decay of the loss) could be interpreted that the BR method with a relatively low (upepsilon) could give a benefit to fitting the solution’s estimation in higher orders due to enrichment of the basis.
({H}_{2}) semi-norm comparison for different versions of pretraining models according to the numbers of hidden units in (x)-axis.
({H}_{2}) semi-norm comparison for different numbers of hidden units according to different versions of pretraining models in (x)-axis.
Note that the computation of the ({H}_{2}) semi-norm on the domain is conducted as follows.
Let us define the output of the activation of the hidden units as ({{hbar }}_{k}) such that
where ({w}_{k}in {mathbb{R}}^{Htimes left(d+1right)}) denotes learnable weights defined in Eq. (1) and the subscript (k) denotes the index of subdomains. The discrete form of the second-order derivative is straightforward as shown in Eq. (20):
where, for a positive real number (delta >0), ({xi }_{i,-delta }) and ({xi }_{i,+delta }) are defined as varying the (i)th component of (xi) by (pm delta), i.e., ({xi }^{i,-delta }=(xi (1),xi (2),dots ,xi (i)-delta ,dots ,xi (d))), ({xi }^{i,+delta }=(xi left(1right),xi left(2right),cdots ,xi left(iright)+delta ,dots ,xi (d))).
The measure of the ({H}_{2}) semi-norm of ({{hbar }}_{k}), denoted ({left|{{hbar }}_{k}right|}_{2}), is defined by the following:
We summed all individual ({left|{{hbar }}_{k}right|}_{2}) so that we could have ({sum }_{k}{left|{{hbar }}_{k}right|}_{2}) for values listed in Table 2.
If the BR method can be effectively applied to the estimation of solutions of PDEs, it stands to reason that it may also be effectively applied to a general task of machine learning with ANNs. To generally prove that the BR method can enhance the performance of ANNs as a tool for machine learning, we tested the BR method on the problem of the differentiation of the exclusive OR (XOR) logic gate, a well-known low-dimensional problem of machine learning on which to apply an ANN. To arrange the problem of XOR in a way that could fit the concept of the BR method, we reorganized the dataset in the computational domain (Omega subset {mathbb{R}}^{H}), where (Omega ={left[0, 1right]}^{2}), as follows (see Fig. 15).
The domain of the XOR problem, data points with labels in red letters (left), and the domain structure to apply the BR method with (upepsilon) (right).
The XOR dataset is organized as (0,0), (1/2, 0), (0, 1/2), (1/2, 1/2), with labels of 1, 0, 0, 1, respectively. We defined the domain ({Omega }_{q}) as ({Omega }_{q}={left[0, 1/2right]}^{2}) which included the dataset (see Fig. 15) and denoted (widehat{{Omega }_{q}}) as (widehat{{Omega }_{q}}) = [(0,0), (1/2, 0), (0, 1/2), (1/2, 1/2)] with the set of labels Y = [1, 0, 0, 1]. Note that the definition of (widehat{{Omega }_{q}^{upepsilon }}) is given in Eq. (15).
Now, we can employ the BR scheme which is illustrated in Fig. 16. Since BR is a pretraining scheme, we determined the efficiency as a function of the number of epochs in the pretraining process. In our experiments, we set the number of pretraining epochs to be 500, 1000, 2000, and 3000. Results are compared with the ordinary training scheme which does not apply any pretraining method. We discretized each subdomain of (Omega) except (widehat{{Omega }_{q}}) into 21 equi-spatial nodes along each axis so that there were about 400 nodes in total.
The BR method applied to the XOR problem.
In Fig. 17, results of the experiments for BR with varying (upepsilon) are shown for the case with 4 hidden units. The values of (upepsilon) used were 0.05, 0.1, 0.2, 0.3, and 0.4. In the plots, data for 500, 1000, 2000, and 3000 epochs of pretraining are indicated as PT500, PT1000, PT2000, and PT3000, respectively. The ordinary training scheme which had no pretraining process was denoted as PT0. Note that initial values of neural weights were set to be the same for each case since we aimed to verify the progression as we increased epochs of the pretraining of each method. Our results revealed that as the number of epochs of pretraining increased, the convergence of the loss to 0 was accelerated. As the size of (upepsilon) increased, the speed of convergence was further accelerated. However, for the case of PT0, the convergence progress appeared to stop in the middle and the training failed to optimize the loss.
Results for the BR method with various sizes of (upepsilon) compared to the ordinary training scheme for the case with 4-hidden units.
To make a comparison in the case of a larger hidden layer, we repeated the experiments with 200 hidden units and plotted the results for ϵ = 0.05, 0.1, and 0.2 as shown in Fig. 18. Similar to previous results with 4-hidden units, as the number of pretraining epochs increased, the convergence of the loss accelerated, although the difference in the speed of convergence was insignificant (Fig. 18). Experimental results shown above suggest that the BR technique applied in the pretraining seems to help the ANN model estimate the solution to the XOR problem. Furthermore, the finding that the convergence speed increased with the number of epochs in the pretraining showed us that the training results might not be caused by a random phenomenon with some vague factors and that the BR method strongly contributed to the effective utilization of the optimization power of the hidden units.
Results of the BR method with various sizes of (upepsilon) for the case of 200-hidden units.
On the other hand, to demonstrate that the basis reconstruction method improved the approximation power of ANNs, the basis function of each hidden unit was developed into a higher state of the degree of polynomials so that the estimation power of the hidden layer could be improved. The approximated function representing the output function of the XOR logic gate given in Fig. 19 shows a wrong approximation of XOR by PT0 on the left and a higher-quality approximation on the right. The solution of the XOR function might have a high order of curvature’s smoothness. Thus, the approximating numerical basis functions of the ANN also need to be in the form of some high-degree polynomials. With prerequisite knowledge, we observed shapes of the hidden units’ pretrained activation. Since the output model of the ANN is given by the linear combination of the hidden units’ activation, a comparison of the hidden units’ activations may give us some fine insight to understand properties of the ANN.
Resulting approximations with 4-hidden units made by methods of PT0 (left) and PT3000 (right).
The estimation’s accuracy using NN depends on the solution’s analytic complexity. If the solution consists of high order polynomials, it needs to employ high order hidden units in NN because the number of hidden units determines the estimation’s order of polynomials. In this study as basic research, we applied the proposed method for a single hidden layered architecture based on the universal approximation theory of the single hidden layered NN25. In our study of the implementation of an ANN-PDE solver, we proposed a pretraining version of ANN-based DDM which used a basis reconstruction technique where weight parameters in the hidden layer were rearranged to enrich the quality of the numerical basis and estimated solutions. The BR method was implemented by setting the output of the ANN to be zero on the neighboring subdomains. The main idea is that, as the ANN is nullified on other subdomains, eventually the core degrees of polynomials constituting the hidden units’ numerical basis might be rearranged to center on their main subdomains. Experiments were conducted with one- and two-dimensional Poisson’s equations for the XOR problem. When the BR method was applied to PDE problems, it was observed that the BR method enhanced the computing power to accelerate the rate at which poor estimations of an ANN on single domains were improved in the DDM model by reducing the computational costs relative to single-domain models. On the other hand, to numerically demonstrate the basis reconstruction property and benefits for the weight parameter’s initialization, we compared the after-pretraining status of the hidden units’ activation in the XOR problem. Results showed that degrees of polynomials of activations by the BR method, understood by measuring the curvature, were enriched. This gives a superior approximation compared with ordinary training procedures.
In this work of developing DDM in the structure of ANNs, the basis reconstructing treatment was effective for estimating solutions to PDEs. It could also be applied well to general machine learning tasks. Thus, it may be applicable as a tool for subjects such as regularization, weight initialization, and speeding computations in training.
Nowadays, lots of research have been conducted applying more deeper architectures. For instance, researchers have used ResNet42 which provides more promising performances than any basic deep neural network (DNN)-based architecture due to the skip-connection’s property to find optimal number of hidden layers without too laborious tests made for some normal DNN architectures. However, for a deeper architecture of neural network, we have to apply the proposed method and test it to know how effective it will be in future works.
Smith, G. D. Numerical Solution of Partial Differential Equations: Finite Difference Methods (Clarendon Press, 1978).
Google Scholar
Hughes, T. J. R. The Finite Element Method (Prentice-Hall, 1987).
MATH Google Scholar
Quarteroni, A. Numerical Approximation of Partial Differential Equation (Springer, 1994).
Book Google Scholar
Minsky, M. Steps toward artificial intelligence. IEEE Proc. IRE 49(1), 8–30 (1961).
MathSciNet Article Google Scholar
Lake, B. M., Salakhutdinov, R. & Tenenbaum, J. B. Human-level concept learning through probabilistic program induction. Science 350, 1332–1338 (2015).
ADS MathSciNet CAS Article Google Scholar
Norouzzadeh, M. S., Nguyen, A., Kosmala, M., Swanson, A., Packer, C. & Clune, J. Automatically identifying wild animals in camera-trap images with deep learning. http://arXiv.org/1703.05830 (2017).
Seo, J.-K., Baik, S. & Lee, S.-H. Simulating the architecture of a termite incipient nest using a convolutional neural network. Ecol. Inform. 44, 94–100 (2018).
Article Google Scholar
Lee, J., Davari, H., Singh, J. & Pandhare, V. Industrial artificial intelligence for industry 4.0-based manufacturing systems. Manuf. Lett. 18, 20–23 (2018).
Article Google Scholar
Nam, M.-D. & Thanh, T.-C. Numerical solution of differential equations using multiquadric radial basis function networks. Neural Netw. 14(2), 185–199 (2001).
Article Google Scholar
Hangelbroek, T. & Ron, A. Nonlinear approximation using Gaussian kernels. J. Funct. Anal. 259(1), 203–219 (2010).
MathSciNet Article Google Scholar
Mehrkanoon, S. & Suykens, J. A. K. Learning solutions to partial differential equations using LS-SVM. Neurocomputing 159, 105–116 (2015).
Article Google Scholar
Sirignano, J. & Spiliopoulos, K. DGM: A deep learning algorithm for solving partial differential equations. http://arXiv.org/1708.07469 (2018).
Chen, X., Duan, J. & Karniadakis, G. E. Learning and meta-learning of stochastic advection-diffusion-reaction system from sparse measurements. Eur. J. Appl. Math. 32, 397–420 (2021).
MathSciNet Article Google Scholar
J. Tompson, J., Schlachter, K., Sprechmann, P. & Perlin, K. Accelerating Eulerian fluid simulation with convolutional networks. http://arXiv.org/1607.03597 (2017).
Li, W., Bazant, M. Z. & Zhu, J. A physics-guided neural network framework for elastic plates: Comparison of governing equations-based and energy-based approaches. Comput. Methods Appl. Mech. Eng. 383, 113933 (2021).
ADS MathSciNet Article Google Scholar
Li, W., Zhu, J., Xia, Y., Gorji, M. B. & Wierzbicki, T. Data-driven safety envelope of lithium-ion batteries for electric vehicles. Joule 3, 2703. https://doi.org/10.1016/j.joule.2019.07.026 (2019).
CAS Article Google Scholar
Han, J., Jentzen, A. & Weinan, E. Solving high-dimensional partial differential equations using deep learning. Proc. Natl. Acad. Sci. U.S.A. 115(34), 8505–8510 (2018).
ADS MathSciNet CAS Article Google Scholar
Raissi, M., Perdikaris, P. & Karniadakis, G. E. Physics-informed neural networks: A deep learning framework for solving forward and inverse problems involving nonlinear partial differential equations. J. Comput. Phys. 378, 686–707 (2019).
ADS MathSciNet Article Google Scholar
Beidokhti, R. S. & Malek, A. Solving initial-boundary value problems for systems of partial differential equations using neural networks and optimization techniques. J. Franklin Inst. 346, 1–11 (2009).
MathSciNet Article Google Scholar
Cheng, T., Lewis, F. L. & Abu-Khalaf, M. Fixed-final-time-constrained optimal control of nonlinear systems using neural network HJB approach. IEEE Trans. Neural Netw. 18(6), 1725–1737 (2007).
Article Google Scholar
Shirvany, Y., Hayati, M. & Moradian, R. Multilayer perceptron neural networks with novel unsupervised training method for numerical solution of the partial differential equations. Appl. Soft Comput. 9(1), 20–29 (2009).
Article Google Scholar
Balam, R. I., Hernandez-Lopez, F., Trejo-Sánchez, J. & Zapata, M. U. An immersed boundary neural network for solving elliptic equations with singular forces on arbitrary domains. Math. Biosci. Eng. 18(1), 22–56 (2021).
MathSciNet Article Google Scholar
Rudd, K., Muro, G. D. & Ferrari, S. A constrained backpropagation approach for the adaptive solution of partial differential equations. IEEE Trans. Neural Netw. Learn. Syst. 25(3), 571–584 (2014).
Article Google Scholar
Jagtap, A. D., Kharazmi, E. & Karniadakis, G. E. Conservative physics-informed neural networks on discrete domains for conservation laws: Applications to forward and inverse problems. Comput. Methods Appl. Mech. Eng. 365, 113028 (2020).
ADS MathSciNet Article Google Scholar
Lagaris, I. E., Likas, A. & Fotiadis, D. I. Artificial neural networks for solving ordinary and partial differential equations. IEEE Trans. Neural Netw. 9, 987–1000 (1998).
CAS Article Google Scholar
Kharazmi, E., Zhang, Z. & Karniadakis, G. E. M. hp-VPINNs: Variational physics-informed neural networks with domain decomposition. Comput. Methods Appl. Mech. Eng. 374, 113547 (2021).
ADS MathSciNet Article Google Scholar
Srivastava, N., Hinton, G., Krizhevsky, A., Sutskever, I. & Salakhutdinov, R. Dropout: A simple way to prevent neural networks from overfitting. J. Mach. Learn. Res. 15(56), 1929–1958 (2014).
MathSciNet MATH Google Scholar
Hornik, K., Stinchcombe, M. & White, H. Multilayer feedforward networks are universal approximators. Neural Netw. 2, 359–366 (1989).
Article Google Scholar
He, K., Zhang, X., Ren, S. & Sun, J. Deep residual learning for image recognition. http://arXiv.org/1512.03385 (2015).
McFall, K. S. & Mahan, J. R. Artificial neural network method for solution of boundary value problems with exact satisfaction of arbitrary boundary conditions. IEEE Trans. Neural Netw. 20(8), 1221–1233 (2009).
Article Google Scholar
Lagaris, I. E., Likas, A. & Fotiadis, D. I. Artificial neural network methods in quantum mechanics. Comput. Phys. Commun. 104, 1–14 (1997).
ADS CAS Article Google Scholar
Leshno, M., Lin, V., Pinkus, A. & Schocken, S. Multilayer feedforward networks with nonpolynomial activation function can approximate any function. Neural Netw. 6, 861–867 (1993).
Article Google Scholar
Lagaris, I. E., Likas, A. C. & Papageorgiou, D. G. Neural-network methods for boundary value problems with irregular boundaries. IEEE Trans. Neural Netw. 11(5), 1041–1049 (2000).
CAS Article Google Scholar
Williams, R. J., Hinton, G. E. & Rumelhart, D. E. Learning representations by back-propagating errors. Nature 323(6088), 533–536 (1986).
ADS Article Google Scholar
Burkardt, J., Gunzburger, M. & Lee, H.-C. POD and CVT-based reduced-order modeling of Navier-Stokes flows. Comput. Methods Appl. Mech. Eng. 196, 337–355 (2006).
ADS MathSciNet Article Google Scholar
Seo, J.-K. & Shin, B.-C. Numerical solutions of Burgers equation by reduced-order modeling based on pseudo-spectral collocation method. J. KSIAM 19(2), 123–135 (2015).
MathSciNet MATH Google Scholar
Cai, X.-C. & Widlund, O. B. Domain decomposition algorithms for indefinite elliptic problems. SIAM J. Sci. Stat. Comput. 13(1), 243–258 (1992).
MathSciNet Article Google Scholar
Toselli, A. & Widlund, O. B. Domain Decomposition Methods—Algorithms and Theory (Springer, 2005).
Book Google Scholar
Funaro, D. Spectral Elements for Transport-Dominated Equations (Lecture Notes in Computational Science and Engineering) (Springer, 1997).
Book Google Scholar
Alipanahi, B., Delong, A., Weirauch, M. T. & Frey, B. J. Predicting the sequence specificities of DNA- and RNA-binding proteins by deep learning. Nat. Biotechnol. 33, 831–838 (2015).
CAS Article Google Scholar
Seo, J.-K. et al. Differentiation of the follicular neoplasm on the gray-scale US by image selection subsampling along with the marginal outline using convolutional neural network. BioMed. Res. Int. 2017, 3098293 (2017).
PubMed PubMed Central Google Scholar
Seo, J.-K. Reduced order modeling of the elliptic systems of the frequency-domain method and its error estimation. Numer. Heat Transf. B Fundam. https://doi.org/10.1080/10407790.2022.2079833 (2022).
Article Google Scholar
Download references
This research was supported by the Institute of Data Science of Korea University, funded by the Korea University’s endowment, Research Project KUIDS-2021, and also supported by Jesus Christ foundation for scientific research of Every Family Mission Korea.
Institute of Data Science, Korea University, 145 Anam-ro, Seongbuk-gu, Seoul, 02841, South Korea
Jeong-Kweon Seo
You can also search for this author in PubMed Google Scholar
J.S. developed the method, performed the numerical experiments, wrote the manuscript.
Correspondence to Jeong-Kweon Seo.
The author declares no competing interests.
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.
Reprints and Permissions
Seo, JK. A pretraining domain decomposition method using artificial neural networks to solve elliptic PDE boundary value problems. Sci Rep 12, 13939 (2022). https://doi.org/10.1038/s41598-022-18315-4
Download citation
Received:
Accepted:
Published:
DOI: https://doi.org/10.1038/s41598-022-18315-4
Anyone you share the following link with will be able to read this content:
Sorry, a shareable link is not currently available for this article.
Provided by the Springer Nature SharedIt content-sharing initiative
By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.
Advertisement
Advanced search
© 2022 Springer Nature Limited
Sign up for the Nature Briefing newsletter — what matters in science, free to your inbox daily.
CCM shares fall 2022 schedule of performances and public events – University of Cincinnati
Audiences are invited to return to the University of Cincinnati College-Conservatory of Music’s (CCM) concert halls and theaters to experience world-class performances and presentations by the next generation of performing and media artists! The college’s fall 2022 schedule of free and ticketed events is now available. Tickets go on sale beginning on Monday, Aug. 22, through the CCM Box Office website.
CCM’s fall lineup of CCM OnStage events includes exemplary faculty and guest artist recitals, dazzling musical theatre, acting, opera and dance productions; exhilarating orchestral, winds, jazz and choral performances, enlightening public lectures — and much more! Learn more about CCM’s schedule of performing and media arts events by referring to the list below or by visiting ccm.uc.edu/onstage.
CCM offers free admission to many of its public events. Some events do require purchased tickets; tickets for those events will go on sale to the general public at 9 a.m. on Monday, Aug. 22. CCMpower members and patrons who purchased subscriptions in 2019-20 were given a Pre-Sale Code, which grants them exclusive access to purchase tickets on Aug. 15-21 before the general on sale date and a 10% discount on ticket prices.
Tickets can be purchased online at ccmonstage.universitytickets.com. The CCM Box Office will re-open to accept orders over the phone at 513-556-4183 and in person at 12:30 p.m. on Monday, Aug. 29.
Join us for a theatrical concert featuring the music of CCM alumnus and Tony Award-winning Broadway composer Stephen Flaherty at 8 p.m. on Friday, Sept. 23. This CCMpower fundraising gala will feature CCM students performing alongside alumni and Broadway performers including Christy Altomare, Ashley Brown, Jason Graae, Aaron Lazar, Bryonha Marie, Stephanie Jae Park, Alton Fitzgerald White and other surprise guests. Tickets go on sale Monday, Aug. 15, 2022. Learn more here.
Be the first to know about future ticket on-sale dates and more: sign up for our email list at ccm.uc.edu/subscribe. CCM will share its spring 2023 schedule of performances and public events in November.
7:30 p.m. Tuesday, August 30
• Faculty Artist Series •
Russell Burge, vibraphone; Sergio Pamies, piano; Aaron Jacobs, bass; and John Taylor, drums
Location: Cohen Family Studio Theater
Admission: FREE
4 p.m. Saturday, Sept. 3
• Guitar Series •
50th ANNIVERSARY GALA CONCERT
Classical Guitar Ensemble
Christopher Wilke, director
Celebrate the 50th anniversary of CCM’s Classical Guitar program with a special concert featuring distinguished alumni in solos and ensembles performed by Amy Brucksch, Duane Corn, Brian DeLay, Murray Holland, Luke Jackson, Julie Goldberg, James Meade, Leighann Narum and Matthew Trkula. The concert will include the world premiere of three new works written for the occasion by Wenbin Lyu, Christopher Teves and Christopher Wilke, as well as a guitar orchestra made of alumni and current students. The gala also recognizes the program’s founder, Professor Emeritus Clare Callahan, with the launch of a new scholarship named in her honor to help promising young guitarists achieve their dreams.
Location: Patricia Corbett Theater
Tickets: $19.50; student and group discounts available.
7:30 p.m. Tuesday, Sept. 6
• CCM String-Quartet-in-Residence •
OPENING CONCERT
Ariel Quartet
Described by the American Record Guide as "a consummate ensemble gifted with utter musicality and remarkable interpretive power," the Ariel Quartet has earned a glowing international reputation. The ensemble is comprised of Alexandra “Sasha” Kazovsky, violin; Amit Even-Tov, cello; Gershon Gerchikov, violin; and Jan Grüning, viola. The group was formed in Israel in 1998 and has served as CCM’s string quartet-in-residence since 2012.
BEETHOVEN: String Quartet in F major, Op. 18, No. 1
BEETHOVEN: String Quartet in G major, Op. 18, No. 2
BEETHOVEN: String Quartet in D major, Op. 18, No. 3
Location: Robert J. Werner Recital Hall
Tickets: $29.50; student and group discounts available.
4 p.m. Sunday, Sept. 11
• Winds Series •
OF THEE WE SING
CCM Wind Symphony and Wind Ensemble
Kevin Michael Holzman and Thomas Gamboa, music directors and conductors
The CCM Wind Symphony and Wind Ensemble perform music by American composers commemorating historical events, places and culture.
Wind Ensemble:
JOLLEY: Motordom
LO PRESTI: Elegy for a Young American
MILBURN: American Hymnsong Suite
NELSON: Savannah River Holiday
Wind Symphony:
ZARE: We Choose to Go to the Moon
DANYEW: Into the Silent Land
WARD (ARR. DRAGON): America the Beautiful
COPLAND: Lincoln Portrait
GOULD: Symphony No. 4, “West Point”
Location: Corbett Auditorium
Tickets: $19.50; student and group discounts available.
7:30 p.m. Sunday, Sept. 11
• Faculty Artist Series •
Dror Biran, piano
Featuring guest artists Paul York, cello; and Geoffrey Herd, violin
Showcasing piano trios by Claude Debussy and Pyotr Ilyich Tchaikovsky.
Location: Robert J. Werner Recital Hall
Admission: FREE
7:30 p.m. Tuesday, Sept. 13
• Faculty Artist Series •
Soyeon Kate Lee, piano
Featuring faculty artist Ran Dank, piano
A recital of works by Meredith Monk, Bach-Kurtag, Marc Andre-Hamelin and Rachmaninoff.
Location: Robert J. Werner Recital Hall
Admission: FREE
7:30 p.m. Wednesday, Sept. 14
• Orchestra Series •
THE OCEAN
CCM Concert Orchestra
Aik Khai Pung, music director and conductor
Featuring faculty artist Quinn Patrick Ankrum, mezzo-soprano
SIBELIUS: The Oceanides, Op. 73
ELGAR: Sea Pictures, Op. 37
FRANCK: Symphony in D Minor
Location: Corbett Auditorium
Admission: FREE
1:30 p.m. Friday, Sept. 16
• The Joseph and Frances Jones Poetker Thinking About Music Lecture Series •
WOMEN AUDIO PRODUCERS AND ENGINEERS, AND THE CASE AGAINST GENDER VENTRILOQUISM
Helen Reddington, Senior Lecturer in Music Production at the University of East London
This talk will focus on women professionals in the UK music industry. Reddington’s The Lost Women of Rock Music (Equinox, 2012) features women punk instrumentalists in the UK in the late 1970s, speaking through the lens of history. These experiences, voiced by the women themselves, challenged the male narrative of the subculture and indeed its very maleness. Reddington’s She’s at the Controls (Equinox, 2019) follows the same methodology, this time focusing on much more recent accounts by women in the music industry. It focuses both on the reasons why it is important that women have equality of access to the music industries, and the impact of often invisible male gatekeeping on the end product (the music) that we listen to and map our lives to. Reddington will discuss issues raised by the women producers and engineers that she interviewed and will contextualize these within the central concept of the book: that of gender ventriloquism.
Location: Online via Zoom. Contact Professor Jeongwon Joe for details.
Admission: FREE
7:30 p.m. Friday, Sept. 16
• Orchestra and Choral Series •
HOPE, FATE, TRIUMPH
CCM Philharmonia, Chamber Choir, Chorale and UC Choruses
Mark Gibson, music director and conductor
Joe Miller, music director
Brett Scott, Reina Dickey and Isiah Maxey, conductors
The CCM Philharmonia opens its 2022-23 season with former Guggenheim Fellow Nkeiru Okoye’s Voices Shouting Out, which the composer wrote in response to 9/11. In her words, “it was a march to acknowledge those fighting on behalf of our safety, and yet a sparkling celebration of life for those who continue living.” Then CCM choirs and UC Choruses join the orchestra for Brahms’ Schicksalslied (Song of Destiny) and Symphony No. 1 in C Minor, Op. 68.
NKEIRU OKOYE: Voices Shouting Out (2002)
BRAHMS: Schicksalslied, Op. 54
BRAHMS: Symphony No. 1 in C Minor, Op. 68
Location: Corbett Auditorium
Tickets: $25; student and group discounts available.
7 p.m. Sunday, Sept. 18
• Jazz Series •
DON’T GIT SASSY: THE MUSIC OF THAD JONES
CCM Jazz Orchestra
Scott Belck, music director
Thad Jones’ big band music embodies the height of sophistication, the depths of hard driving swing, the relentless groove of the blues, and the beauty of a compositional style like no other. Join us for an evening celebrating the swinging music of this American jazz master.
Location: Corbett Auditorium
Tickets: $25; student and group discounts available.
7:30 p.m. Tuesday, Sept. 20
• Faculty Artist Series •
Pavel Vinnitsky, clarinet; and Ilya Finkelshteyn, cello
Featuring guest artist Anna Vinnitsky, piano
WEBER: Grand Duo concertant, Op. 48
A. VINNITSKY: Spring of 2020: Four Months Without Words for Clarinet and Piano
BRAHMS: Trio in A Minor for Clarinet, Cello and Piano, Op. 114
Location: Robert J. Werner Recital Hall
Admission: FREE
8 p.m. Friday, Sept. 23
• CCMpower Special Event Series •
30 YEARS OF STEPHEN: THE MUSIC OF STEPHEN FLAHERTY
CCMpower Fundraising Gala
Join us for the world premiere of a theatrical concert featuring works by CCM alumnus and Broadway composer Stephen Flaherty. The performance features CCM students with alumni and Broadway stars Christy Altomare, Ashley Brown, Jason Graae, Aaron Lazar, Bryonha Marie, Stephanie Jae Park, Alton Fitzgerald White and other surprise guests. Audiences will also experience the premiere of two never before heard compositions performed by Flaherty himself! Proceeds benefit CCM student scholarships and projects.
Location: Corbett Auditorium
Tickets: Tickets on sale August 15. Special ticket pricing and limited seating. Sponsorship and ticket information available online.
7:30 p.m. Tuesday, Sept. 27
• Jazz Series •
IN THE TRADITION
CCM Jazz Lab Band
Craig Bailey, music director
Featuring swing music from the masters, including Count Basie Thad Jones, Duke Ellington, Bill Holman, Woody Herman and more.
Location: Corbett Auditorium
Tickets: $19.50; student and group discounts available.
7:30 p.m. Tuesday, Sept. 27
• Faculty Artist Series •
Michael Chertock, piano
Featuring faculty artist Dror Biran, piano, and guest artist Charles Morey, violin
A performance of works by Grieg, Rachmaninoff and Fiday.
Location: Robert J. Werner Recital Hall
Admission: FREE
7:30 p.m. Tuesday, Sept. 27
• Faculty Artist Series •
Ric Hordinski, guitar
Featuring guest artists Ted Nelson, cello, and Eric Bates, violin
Original works for six- and 12- string, fretted and fretless, prepared and under-prepared guitar.
Location: Cohen Family Studio Theater
Admission: FREE
8 p.m. Thursday, Sept. 29
8 p.m. Friday, Sept. 30
2 and 8 p.m. Saturday, Oct. 1
2 p.m. Sunday, Oct. 2
• Play Series •
CINCINNATI DIONYSIA:
LYSISTRATA + THE TROJAN WOMEN
Adapted by Ellen McLaughlin
Brant Russell, director
Caitlin Hines, dramaturg and director of movement
Megan Steigerwald-Ille, musicology consultant
Ellen McLaughlin’s adaptations of Aristophanes’ Lysistrata and Euripides’ The Trojan Women tell stories of women impacted by war and their struggle to take control of their fates. A satirical comedy, Lysistrata focuses on women who try to end the Peloponnesian War by going celibate until their men see reason and lay down their arms. Known as the first anti-war play in the western canon, The Trojan Women focuses on the surviving women who grapple with grief, fear and the total devastation of their homes after their city’s defeat in the Trojan War.
Estimated runtime: 2 hours plus an intermission
Location: Patricia Corbett Theater
Tickets: $32.50; student and group discounts available.
7:30 p.m. Thursday, Sept. 29
• Winds Series •
MINIMALIST MASTERS
Musica Nova
Kevin Michael Holzman, music director and conductor
CCM Musica Nova’s inaugural concert features the music of renowned minimalist composers, including Steve Reich’s stunning audio/visual collaboration Reich/Richter.
ADAMS: Son of Chamber Symphony
PÄRT: Fratres
REICH: Reich/Richter
Location: Corbett Auditorium
Admission: FREE
1:30 p.m. Friday, Sept. 30
• The Joseph and Frances Jones Poetker Thinking About Music Lecture Series •
HIDING IN PLAIN SIGHT: VARIETIES OF INVERTIBLE CANON IN THE LONG SIXTEENTH CENTURY
Denis Collins, Associate Professor in Musicology from University of Queensland, Australia
This presentation takes as its starting point a seeming error in a musical example in Gioseffo Zarlino’s Le istitutioni harmoniche (1558). Careful comparison with similar examples shows that all of his examples are correct as notated. They demonstrate a contrapuntal technique that results in a type of invertible canon in performance. This lecture points to other situations where Zarlino’s theories can help in unravelling the complexities of 16th-century counterpoint. It also discusses how canons, including those that employ complex procedures, enjoyed a long history as title pages or frontispieces to manuscript collections or printed volumes of music. These canons were often presented in geometric shapes whose symbolic significance could be reinforced by accompanying artwork or textual commentaries.
Location: Online via Zoom. Contact Professor Steven Cahn for details.
Admission: FREE
7:30 p.m. Friday, Sept. 30
• Winds Series •
MASTERWORKS
CCM Wind Symphony
Kevin Michael Holzman, music director and conductor
The CCM Wind Symphony performs masterworks by world-renowned composers.
MENNIN: Canzona
GOTO: A Wild Rose Above
LUKÁS: Musica Boema
DICKERSON: Essay for Band
GOTKOVSKY: Symphonie pour grande orchestra d’harmonie
Location: Corbett Auditorium
Tickets: $19.50; student and group discounts available.
8 p.m. Thursday, Sept. 29
8 p.m. Friday, Sept. 30
2 and 8 p.m. Saturday, Oct. 1
2 p.m. Sunday, Oct. 2
• Play Series •
CINCINNATI DIONYSIA:
LYSISTRATA + THE TROJAN WOMEN
Adapted by Ellen McLaughlin
Brant Russell, director
Caitlin Hines, dramaturg and director of movement
Megan Steigerwald-Ille, musicology consultant
Ellen McLaughlin’s adaptations of Aristophanes’ Lysistrata and Euripides’ The Trojan Women tell stories of women impacted by war and their struggle to take control of their fates. A satirical comedy, Lysistrata focuses on women who try to end the Peloponnesian War by going celibate until their men see reason and lay down their arms. Known as the first anti-war play in the western canon, The Trojan Women focuses on the surviving women who grapple with grief, fear and the total devastation of their homes after their city’s defeat in the Trojan War.
Estimated runtime: 2 hours plus an intermission
Location: Patricia Corbett Theater
Tickets: $32.50; student and group discounts available.
5 p.m. Saturday, Oct. 1
• Starling Series •
STARLING SHOWCASE
Kurt Sassmannshaus, music director
Showcasing Starling violin students from the college and pre-college programs.
Location: Robert J. Werner Recital Hall
Admission: FREE
2 p.m. Sunday, Oct. 2
• Faculty Artist Series •
Daniel Weeks, tenor
Donna Loewy, piano
A song recital featuring works by Quilter, Liszt, Hahn, Tosti, Donizetti and Berger.
Location: Robert J. Werner Recital Hall
Admission: FREE
7:30 p.m. Monday, Oct. 3
• Winds Series •
CCM BRASS CHOIR SHOWCASE
Timothy Northcut, music director and conductor
CCM’s nationally recognized Brass Choir performs selections featuring brass ensembles large and small consisting of students from the CCM horn, trumpet, trombone, tuba/euphonium and percussion studios.
STEPHENSON: Fanfare and March
GJEILO: Serenity (O Magnum Mysterium)
HUSA: Divertimento
EWAZEN: Symphony in Brass
…and more from Brass Chamber Ensembles
Location: Corbett Auditorium
Admission: FREE
7:30 p.m. Thursday, Oct. 6
• Winds Series •
INSPIRATIONS
CCM Wind Ensemble
Thomas Gamboa, music director and conductor
SOUSA: Washington Post March
GOULD: Ballad for Band
TOCH: Spiel for Wind Orchestra, Op. 39
VAUGHAN WILLIAMS: English Folk Song Suite
GERSHWIN/ VAN GILS: An American in Paris
WAGNER/BOURGEOIS: Elsa’s Procession to the Cathedral
Location: Corbett Auditorium
Admission: FREE
8 p.m. Thursday, Oct. 6
8 p.m. Friday, Oct. 7
2 and 8 p.m. Saturday, Oct. 8
2 p.m. Sunday, Oct. 9
• Musical Theatre Series •
SHE LOVES ME
Book by Joe Masteroff
Music by Jerry Bock
Lyrics by Sheldon Harnick
Based on a play by Miklos Laszio
Originally Directed on Broadway by Harold Prince
Originally Produced on Broadway by Harold Prince
In Association with Lawrence N. Kasha and Philip C. McKenna
Original Orchestrations by Don Walker
Adapted by Frank Matosich, Jr.
Eric Byrd, director and choreographer
Ian Axness, musical director
Considered by many to be the most charming musical ever written, She Loves Me is a warm romantic comedy with an endearing innocence and a touch of old-world elegance. Based on a play that inspired the beloved film You’ve Got Mail, the musical shares an enemies-to-lovers story about two shop clerks who normally don’t get along — until they begin an anonymous love letter exchange after they respond to a lonely-hearts ad in the newspaper.
Estimated runtime: 2 hours and 30 minutes
Location: Cohen Family Studio Theater
Tickets: $29.50; student and group discounts available.
*Please note that traffic and parking around campus will be heavier than normal on Oct. 8 due to UC Homecoming Activities. CCM encourages patrons to purchase parking in advance through the Box Office.
2 p.m. Sunday, Oct. 9
• Faculty Artist Series •
Christopher Wilke, guitar
Featuring faculty artists Quinn Patrick Ankrum, mezzo-soprano; James Bunte, saxophone; and Michael Unger, harpsichord
Music for theorbo, lute, classical guitar and electric guitar featuring works by Benjamin Britten, Miguel Llobet, Dorothy Chang and others.
Location: Robert J. Werner Recital Hall
Admission: FREE
7:30 p.m. Saturday, Oct. 15
• Orchestra Series •
FULL FORCE
CCM Philharmonia
Mark Gibson, music director and conductor
Join us for a performance of Aya Yoshida’s Double Face, winner of CCM’s Alexander Zemlinsky Prize for Composition in 2019, and Sergei Rachmaninoff’s Symphony No. 2, known for its rich orchestration and passionate melodies.
AYA YOSHIDA: Double Face (2015; rev. 2022)
RACHMANINOFF: Symphony No. 2, Op. 27 (1907)
Location: Corbett Auditorium
Tickets: $25; student and group discounts available.
7:30 p.m. Tuesday, Oct. 18
• Percussion Series •
CCM Percussion Ensemble
James Culley, music coordinator
Location: Cohen Family Studio Theater
Admission: FREE
7:30 p.m. Wednesday, Oct. 19
• Composition Series •
SONIC EXPLORATIONS
Mara Helmuth, music director
An evening of electroacoustic music by CCM students, faculty and guests.
Location: Cohen Family Studio Theater
Admission: FREE
7:30 p.m. Wednesday, Oct. 19
• Guest Artist Series •
Palmetto Saxophone Quartet
Featuring the premiere of a new work by CCM Saxophone Instructor Om Srivastava.
Location: Robert J. Werner Recital Hall
Admission: FREE
8 p.m. Thursday, Oct. 20
8 p.m. Friday, Oct, 21
2 and 8 p.m. Saturday, Oct. 22
2 p.m. Sunday, Oct. 23
8 p.m. Thursday, Oct. 27
8 p.m. Friday, Oct, 28
2 and 8 p.m. Saturday, Oct. 29
2 p.m. Sunday, Oct. 30
• Musical Theatre Series •
SOMETHING ROTTEN
Book by Karey Kirkpatrick and John O’Farrell
Music and Lyrics by Wayne Kirkpatrick and Karey Kirkpatrick
Conceived by Karey Kirkpatrick and Wayne Kirkpatrick
Diane Lala, director and choreographer
Matthew J.C. Welsh, musical director
Two brothers set out to write the world’s first musical in this hilarious mash-up of 16th-century Shakespeare and 21st-century Broadway. Nick and Nigel Bottom are desperate to write a hit play but are stuck in the shadow of that Renaissance rock star known as "The Bard." When a local soothsayer foretells that the future of theatre involves singing, dancing and acting at the same time, Nick and Nigel write the world’s very first musical. But amidst the scandalous excitement of opening night, the Bottom Brothers realize that reaching the top means being true to thine own self, and all that jazz.
Estimated runtime: 2 hours and 10 minutes
Location: Patricia Corbett Theater
Tickets: $39.50; student and group discounts available.
7:30 p.m. Friday, Oct. 21
• Orchestra Series •
POLISH ARTISTRY
CCM Concert Orchestra
Aik Khai Pung, music director and conductor
Featuring faculty artist Rick VanMatre, saxophone
Join us for a rare collaboration of Polish Artists. Anna VanMatre’s spectacular artworks inspired Piotr Szewczyk to compose a new saxophone concerto that will be premiered by Rick VanMatre.
LUTOSŁAWSKI: Symphonic Variations
SZEWCZYK: Concerto for Saxophone (To Those Who Protect) *World premiere
STRAVINSKY: Firebird Suite (1919 version)
Location: Corbett Auditorium
Admission: FREE
7 p.m. Saturday, Oct. 22
• Starling Series •
Starling Chamber Orchestra
Kurt Sassmannshaus, music director
Showcasing the superbly talented young students from the Starling Preparatory String Project.
Location: Robert J. Werner Recital Hall
Admission: FREE
7 p.m. Sunday, Oct. 23
• Jazz Series •
A NIGHT WITH TAMIR HENDELMAN
CCM Jazz Orchestra
Scott Belck, music director
Featuring guest artist Tamir Hendelman, piano
Award-winning jazz pianist Tamir Hendelman joins the CCM Jazz Orchestra for a night of classic swinging piano. Hendelman is a shining star on the jazz scene, a true virtuoso, he has performed with the Jeff Hamilton Trio, the Clayton-Hamilton Jazz Orchestra, Harry Allen, Teddy Edwards, Warren Vache, Houston Person, Jeff Clayton, Nick Brignola, Phil Upchurch, Rickey Woodard, John Clayton and Barbara Morrison.
Location: Corbett Auditorium
Tickets: $25; student and group discounts available.
7:30 p.m. Tuesday, Oct. 25
• Composition Series •
A VIEW FROM THE EDGE
A showcase of newly-written musical works by CCM’s brilliant composition majors.
Location: Cohen Family Studio Theater
Admission: FREE
7:30 p.m. Tuesday, Oct. 25
• CCM String-Quartet-in-Residence •
BARTOK AND SCHUBERT
Ariel Quartet
CCM’s internationally acclaimed String-Quartet-in-Residence continues its concert series with works by Béla Bartók and Franz Schubert. Specific repertoire TBA.
Location: Robert J. Werner Recital Hall
Tickets: $29.50; student and group discounts available.
7 p.m. Saturday, Oct. 29
• Choral Series •
LIFE AND DEATH: BACH DANCES
CCM Chamber Choir
Joe Miller, music director and conductor
Featuring faculty artist Elliot Madore, baritone
Featuring guest choreographer Taylor Carrasco and guest artists from Cincinnati Ballet
J.S. Bach’s Magnificat in D Major and Ich habe genung come to life with dance and theatrical staging at Cincinnati’s Christ Church Cathedral. The performance of the combined Baroque masterpieces features a collaboration with guest artists from the Cincinnati Ballet, professional musicians and CCM’s faculty and student artists.
Location: Christ Church Cathedral, 318 E 4th St, Cincinnati, OH 45202
Tickets: $19.50; student and group discounts available. Tickets will not be available to purchase in-person at the venue.
2 p.m. Sunday, Oct. 30
• CCM Prep Series •
GINO DIMARIO SCHOLARSHIP RECITAL
CCM Prep music students who are candidates to receive the fall 2022 Gino DiMario Memorial Scholarship will perform at this event along with past DiMario scholars. Scholarship winners will be announced following the performances.
Location: Robert J. Werner Recital Hall
Admission: FREE
7:30 p.m. Tuesday, Nov. 1
• Guitar Series •
Classical Guitar Ensemble
Christopher Wilke, director
Solos, duos, trios and quartets for classical guitar.
Location: Robert J. Werner Recital Hall
Admission: FREE
7:30 p.m. Wednesday, Nov. 2
• Choral Series •
UC CHORUSES FALL CONCERT
Alta Petit, Juncta Juvant and Cabaret Singers
Reina Dickey, Isiah Maxey and Matthew Swope, music directors and conductors
Comprised of students from all 14 of UC’s colleges, the UC’s Alta Petit Treble Chorus, Juncta Juvant Tenor/Bass Chorus and Cabaret Singers join together for a diverse musical program of texts, time periods and composers.
Location: Corbett Auditorium
Admission: FREE
8 p.m. Thursday, Nov. 3
8 p.m. Friday, Nov. 4
2 and 8 p.m. Saturday, Nov. 5
2 p.m. Sunday, Nov. 6
• Play Series •
A DREAM PLAY
By August Strindberg
Translated by Harry G. Carlson
Vincent DeGeorge, director
A young woman from another world is sent to earth to see what it is like to be human. In Strindberg’s A Dream Play, written in 1901, the playwright has sought to imitate the disconnected but apparently logical form of a dream. Anything can happen, everything is possible and plausible. Time and space do not exist. Using the backdrop of real-life events, Strindberg spins new patterns of memories, experiences, inventions and absurdities all in the quest to discover if life is really worth living.
Estimated runtime: 100 minutes, no intermission
Location: Cohen Family Studio Theater
Tickets: $27; student and group discounts available.
*Please note that traffic and parking around campus will be heavier than normal on Nov. 5 due to a UC Football game. CCM encourages patrons to purchase parking in advance through the Box Office.
7:30 p.m. Thursday, Nov. 3
• Jazz Series •
NEW YORK
CCM Jazz Lab Band
Craig Bailey, music director
Featuring compositions and arrangements by artists who have been influenced by the New York scene and lifestyle.
Location: Patricia Corbett Theater
Tickets: $19.50; student and group discounts available.
7:30 p.m. Friday, Nov. 4
• Winds Series •
DREAMSCAPE
CCM Wind Symphony
Featuring the CCM Youth Wind Ensemble (CYWE)
Kevin Michael Holzman, music director and conductor
Featuring faculty artist Denise Tryon, horn
Featuring guest composer Catherine Likhuta
Join the CCM Wind Symphony and Youth Wind Ensemble as we explore the magical world of dreams. Additional repertoire to be announced.
Wind Symphony:
THEOFANIDIS: I wander the world in a dream of my own making
BARFIELD: Here We Rest
LIKHUTA: Vivid Dreams Concerto for Low Horn and Wind Ensemble
MASLANKA: A Child’s Garden of Dreams
Location: Corbett Auditorium
Tickets: $19.50; student and group discounts available.
7:30 p.m. Wednesday, Nov. 9
• Choral Series •
CCM Chorale
Brett Scott, music director and conductor
CCM Chorale presents a concert of newly composed music by a diverse group of composers from around the world, including several North American premieres.
Location: Cohen Family Studio Theater
Admission: FREE
7:30 p.m. Thursday, Nov. 10
• Winds Series •
THROUGH THE LOOKING GLASS
Musica Nova
Kevin Michael Holzman, music director and conductor
Featuring Michael Delfín, harpsichord
CCM Musica Nova Presents a program of contemporary compositions in homage to past eras, composers and compositions, featuring CCM Alumnus Evan Williams’ Dead White Man Music.
STUCKY: Partita-Pastorale, after J.S.B.
LASH: Subtilior, Lamento
THORVALDSDOTTIR: Hrím
WILLIAMS: Dead White Man Music
Location: Patricia Corbett Theater
Admission: FREE
7:30 p.m. Saturday, Nov. 12
• Orchestra Series •
MOZART AND MORE
CCM Philharmonia
Mark Gibson, music director and conductor
CCM Philharmonia performs W.A. Mozart’s Violin Concerto No. 5, featuring the student winner of CCM’s Violin Concerto Competition. Additional repertoire to be announced.
W.A. MOZART: Violin Concerto No. 5 in A Major, K. 219
Location: Patricia Corbett Theater
Tickets: $25; student and group discounts available.
7:30 p.m. Tuesday, Nov. 15
• Winds Series •
THE PLACES YOU’LL GO
CCM Wind Ensemble
Thomas Gamboa, music director and conductor
Featuring the CCM Youth Wind Ensemble (CYWE) Symphonic Winds
ITO: Festal Scenes
RESPIGHI: Huntingtower Ballad
CHÁVEZ/ERICKSON: Sinfonia India
DELLO JOIO: Scenes from the Louvre
VILLA LOBOS/REED: Bachianas Brasileiras
Location: Patricia Corbett Theater
Tickets: $19.50; student and group discounts available.
8 p.m. Friday, Nov. 18
8 p.m. Saturday, Nov. 19
2 p.m. Sunday, Nov. 20
• Opera Series •
DIALOGUES OF THE CARMELITES
Music and Libretto by Frances Poulenc
Libretto by Emmet Lavery
TBA, guest conductor
Robin Guarino, director
Based on a true story, this haunting opera tells the tale of the Carmelite Nuns, a small order put to death during the French Revolution. The fictional Parisian aristocrat Blanche de la Force attempts to find refuge with the nuns, setting the stage for a psychological study of devotion in the face of terror. Sung in French with English supertitles.
Estimated runtime: 3 hours and 6 minutes with intermission
Location: Corbett Auditorium
Tickets: $35.50-$39.50; student and group discounts available.
2 p.m. Saturday, Nov. 19
• Guest Artist Series •
Hinge Quartet
Location: Cohen Family Studio Theater
Admission: FREE
4 p.m. Sunday, Nov. 20
• Jazz Series •
JAZZ HOLIDAY EXTRAVAGANZA
CCM Jazz Orchestra
Scott Belck, music director
Celebrate the holidays with the hot rhythms and cool sounds of Ella Fitzgerald, Frank Sinatra, Stan Kenton, Count Basie and Duke Ellington, and a few surprise guest vocalists who will help us swing into the season!
Location: Patricia Corbett Theater
Tickets: $29.50; student and group discounts available.
7:30 p.m. Monday, Nov. 21
• Winds Series •
CCM BRASS CHOIR SHOWCASE
Timothy Northcut, music director and conductor
CCM’s nationally recognized Brass Choir performs classical, popular and folk selections, featuring brass ensembles large and small consisting of students from the CCM horn, trumpet, trombone, tuba/euphonium and percussion studios.
D’ADDONA: Grand Fanfare
PRICE: Adoration
PLOG: Symphony for Brass
…and more from Brass Chamber Ensembles
Location: Corbett Auditorium
Admission: FREE
7:30 p.m. Tuesday, Nov. 22
• Orchestra Series •
KODÁLY AND DVOŘÁK
CCM Concert Orchestra
Aik Khai Pung, music director and hammered dulcimer
Featuring alumni guest artist Jiannan Cheng, conductor
KODÁLY: Háry János Suite
DVOŘÁK: Symphony No. 9 in E Minor, Op. 95 “From the New World”
Location: Corbett Auditorium
Admission: FREE
7:30 p.m. Wednesday, Nov. 30
• Orchestra Series •
VOCAL CHAMBER SHOWCASE
Musica Nova
Gwen Coleman, music director and conductor
Join us for a music of delightful music for chamber orchestra and voice, performed by CCM student artists.
HÄNDEL: Let the Bright Seraphim Eternal Source of Light Divine
RAVEL: 5 Mélodies populaires grecques
ROREM: Selections from Aftermath
SHENG: Three Love Songs
Location: Robert J. Werner Recital Hall
Admission: FREE
7:30 p.m. Thursday, Dec. 1
• Chamber Music Series •
CCM CHAMBER CONCERT: PIANO
Sarah Minneman, chamber music coordinator
Featuring CCM student artists performing chamber music pieces for piano.
Location: Mary Emery Hall, Room 3250
Admission: FREE
8 p.m. Thursday, Dec. 1
8 p.m. Friday, Dec. 2
8 p.m. Saturday, Dec. 3
3 p.m. Sunday, Dec. 4
• Dance Series •
DANCE WORKS
Shauna Steele, director
A mixed repertoire performance featuring ballet and contemporary dance, including a holiday selection in collaboration with CCM’s costume design and technology students.
Location: Patricia Corbett Theater
Tickets: $29.50; student and group discounts available.
2 and 5 p.m. Saturday, Dec. 3
2 and 5 p.m. Sunday, Dec. 4
FEAST OF CAROLS
CCM Chamber Choir, Chorale and Concert Orchestra, UC Choruses and Cincinnati Youth Choir
Featuring guest choirs from Sycamore High School, Walnut Hills High School and Conner High School
Joe Miller, Brett Scott, Aik Khai Pung, Robyn Lana, Reina Dickey and Isiah Maxey, conductors with guest conductors Ken Holdt, Antony Nims and John DeFerraro
Bring family and friends to CCM to welcome the 2022 holiday season with festive choral favorites performed by CCM’s, UC’s and CYC’s fabulous choirs and outstanding guest choirs.
Location: Corbett Auditorium
Tickets: $21.50-$25.50; student and group discounts available.
7 p.m. Monday, Dec. 5
• CCM Prep Series •
CCM Youth Wind Ensemble and CYWE Chamber Ensembles
Kevin Michael Holzman, music director and conductor
James Bunte, Thomas Gamboa and Michael Mergen, chamber ensemble directors
The area’s most talented high school instrumentalists perform traditional and contemporary band music.
Location: Corbett Auditorium
Admission: FREE
7 p.m. Tuesday, Dec. 6
• CCM Prep Series •
CCM Jr. Youth Wind Ensemble and CYWE Symphonic Band
Rick Canter and Christopher Nichter, conductors
The area’s most talented middle and high school instrumentalists perform traditional and contemporary band music.
Location: Corbett Auditorium
Admission: FREE
7:30 p.m. Wednesday, Dec. 7
UC Symphony Orchestra
Brendan Boyle, music director and conductor
Comprised of non-music majors, UC’s campus orchestra is designed to provide students with an opportunity to share their love of great music from the orchestral repertoire.
Location: Corbett Auditorium
Admission: FREE
7:30 p.m. Thursday, Dec. 8
UC Bearcat Bands
Christopher Nichter, music director and conductor
Join the UC Bearcat Bands for a thrilling evening of music. Consisting primarily of students majoring in fields outside music, these future musical citizens are sure to impress and inspire!
Location: Corbett Auditorium
Admission: FREE
7:30 p.m. Friday, Dec. 9
7:30 p.m. Saturday, Dec. 10
2 p.m. Sunday, Dec. 11
• CCM Prep Series •
FALL YOUTH BALLET CONCERT
Tricia Sundbeck, director
The CCM Youth Ballet Companies feature talented students from ages nine through adult, performing traditional and contemporary works choreographed by CCM and CCM Prep faculty.
Location: Patricia Corbett Theater
Tickets: Prices start at $19.50; student and group discounts available.
2 p.m. and 5 p.m. Saturday, Dec. 10
• CCM Ensemble-in-Residence Series •
HOME FOR THE HOLIDAYS
Cincinnati Youth Choir
Robyn Lana, music director
Join the Cincinnati Youth Choir as children and youth from across the Greater Cincinnati area celebrate the season and the many cultures of the community.
Location: Corbett Auditorium
Tickets: $15; student and group discounts available.
2 p.m. Sunday, Dec. 11
• CCM Prep Series •
CCM Junior Strings
Rachel Bierkan, director
CCM Junior Strings features the area’s premiere string players, ages 10-14, performing a variety of traditional, contemporary and multi-cultural string orchestra music.
Location: Corbett Auditorium
Admission: FREE
All events listed above take place in CCM Village on UC’s campus unless otherwise indicated. Please see individual event information for details and ticket ordering information. Repertoire and dates are subject to change. View CCM’s inclement weather policy here.
All of CCM’s ticketed events will have digital programs instead of printed programs. QR codes will be displayed at events to give access to the digital programs to patrons. The programs will also be accessible through the links below. Programs for events will be shared shortly before the event date.
Tickets are on sale beginning on Monday, Aug. 15, for 2019-20 subscribers; pre-sale information was shared to previous subscribers, who have exclusive access to tickets Aug. 15-21. Tickets go on sale to the general public on Monday, Aug. 22.
Ticket discounts are available for UC and non-UC students and groups.
Our Box Office staff is always ready to answer your questions by email at [email protected]. The Box Office will re-open to phone and in-person purchases on Monday, Aug. 29.
Experiencing the magic of CCM OnStage has never been easier! Purchase your tickets:
CCM OnStage patrons can order prepaid parking at $10 for each performance selected; parking passes are available for purchase through the CCM Box Office. Prepaid parking is date/performance specific. CCM Garage parking rates for a performance or special event is usually available for $10-15.
CCM’s Box Office hours are Mon-Fri from 12:30-6 p.m., Sat. from 12-4 p.m. and one hour prior to curtain for ticketed performances. Hours subject to change based on the University of Cincinnati calendar.
CCM’s schedule of student recitals and other public events can be found online here. Performance dates and details subject to change.
Additional information can be found on UC’s calendar.
Stay up to date on CCM news and events by subscribing to our Next OnStage email newsletter, which is published every other Wednesday. Visit ccm.uc.edu/subscribe or text* CCMONSTAGE to 22828 to sign up for email updates. *Message and data rates may apply.
You can also connect with us on Facebook, Twitter, Instagram, YouTube and LinkedIn.
CCM will share its spring 2023 schedule of performances and public events in November.
A preeminent institution for the performing and media arts, the University of Cincinnati College-Conservatory of Music (CCM) is the largest single source of performing arts presentations in the state of Ohio.
All event dates and programs are subject to change. For a complete calendar of events, please visit us online at ccm.uc.edu.
Featured image at the top: Stage lighting reveals a side view of audience sitting in Corbett Auditorium. Photo/TM Photography.
Curt Whitacre | Director of Marketing/Communications | UC College-Conservatory of Music
[email protected] | 513-556-2683
August 18, 2022
Design World featured the University of Cincinnati's Frank Rybicki in an article about different ways CAD and 3-D printing have helped advance medical care.
August 17, 2022
The University of Cincinnati's Michael Hegener, PharmD, was a panelist on WVXU's Cincinnati Edition discussing the under-regulated dietary supplements industry.
August 17, 2022
During the 2022-23 performance season, the UC College-Conservatory of Music will continue to offer digital programs for all of its free and ticketed events in place of printed programs. These paperless programs will be available on CCM's website, and they can also be accessed via smartphone/mobile device by scanning the Quick Response (QR) code displayed outside of each CCM performance venue.
University of Cincinnati
2600 Clifton Ave.
Cincinnati, OH 45220
513-556-0000
UC Tools
About Us
University of Cincinnati | 2600 Clifton Ave. | Cincinnati, OH 45221 | ph: 513-556-6000
Alerts | Clery and HEOA Notice | Notice of Non-Discrimination | eAccessibility Concern | Privacy Statement | Free Speech | Copyright Information
© 2020 University of Cincinnati
University of Cincinnati
2600 Clifton Ave.
Cincinnati, OH 45220
513-556-0000
© 2020 University of Cincinnati
Police audio of insensitive remarks after officer was shot lead to investigation – WPLG Local 10
The latest local, regional and national news events are presented by the Local 10 News Team along with updated sports, weather and traffic.
The judge, prosecutors and defense attorneys are gathering again Thursday for a hearing in the penalty phase of the Parkland school shooter’s trial.
LIVE
Bridgette Matter, Reporter
Published:
Bridgette Matter, Reporter
MIAMI – A Miami-Dade police officer was critically injured Monday night in a shooting in Miami.
That officer has been identified as 29-year-old Cesar Echaverry.
He is in critical condition and being monitored at Jackson Memorial Hospital.
The suspect who officers were after was wanted in an armed robbery that occurred in Broward County was shot dead.
On Tuesday, a clip of a radio dispatch began going around on social media allegedly of a City of Miami police officer making insensitive remarks as the situation played out Monday night.
Immediately after Officer Echaverry was shot, efforts to get him to a hospital began.
In the audio, an officer can be heard talking about seeing police vehicles “passing my car at 60 mph on 7th Avenue.”
That officer is then advised that what he saw is in reference to an officer down.
“Okay. Let them know they’re going to have another officer down if they going at 70 mph near my car,” the officer replied.
The South Florida Police Benevolent Association says the audio is not a reflection of the City of Miami Police Department.
Steadman Stahl with the South Florida Police Benevolent Association told Local 10 News that it was a Miami police officer heard on the recording.
The audio recordings are under investigation.
Local 10 News also obtained a recording of a supervisor reprimanding the officer who made the remarks.
“Let’s try to use a little bit more tact and not hang anyone up. If we have an officer down, they’re going to pass you at 80 mph next time,” the supervisor said.
City of Miami police released a statement on Tuesday that said, in part:
“If in fact it is determined that it is our officer, he will be held accountable for his actions. It is unsettling and disturbing listening to the lack of concern, compassion, and empathy.”
A Miami police spokesperson told Local 10 News’ Bridgette Matter late Tuesday afternoon they did not know the identity of the officer yet.
Copyright 2022 by WPLG Local10.com – All rights reserved.
A major police response was observed in Miami Springs late Tuesday afternoon.
The man shot and killed by Miami-Dade police officers in a Monday night shootout that left one officer critically wounded was a 32-year-old man from Michigan, sources tell Local 10 News.
The officer has been identified in a department memo as 29-year-old Detective Cesar Echaverry, a member of the department’s Robbery Intervention Detail.
Bridgette Matter joined the Local 10 News team as a reporter in July 2021. Before moving to South Florida, she began her career in South Bend, Indiana and spent six years in Jacksonville as a reporter and weekend anchor.
email
facebook
twitter
instagram
If you need help with the Public File, call (954) 364-2526.
Copyright © 2022 Local10.com is published by WPLG INC., a Berkshire Hathaway company.
Shofar Music Production to host "Moments with Charles Deh" in Sept – Graphic Online
The event aims to provide an opportunity for those in the business environment to appreciate their highly affluent customers, senior management, colleagues and their families on a memorable night of classy live jazz music.
The event has lined up exceptional performers including sensational South African jazz artistes Sinethemba Hlengetwa, Mabongi Fero, and unparalleled, award-winning Ghanaian artist Akwaboah among others.
Speaking at the launch of the event, the founder of Shofar Music Production Charles Deh said, he is very optimistic that the event will go a long way towards enhancing the networking and brand sales for corporate entities.
He took the opportunity to urge lovers of gospel and jazz music not to miss this iconic event.
“I am very glad and grateful to the Almighty God to launch the event which will soon receives a global recognition. It’s my prayer that this important event will strengthen brands and create positive top-of-mind awareness amongst existing and potential clients”, he said.
Charles Deh is a banker by profession and one of Ghana’s most talented pianists with a great deal of skill and experience in music directing, performance, musical talent development and songwriting.
He has played alongside world-renowned Gospel artistes such as Lionel Peterson and Donnie Maclurkin.
He has also worked with Ntokozo Mbambo, Nqubeko Mbatha, Mahalia Buchanan, Khaya Mthetwa, Uche Agu and Patrick Duncan all from Joyous Celebration Fame, renowned Jazz guitarist Kunle Ayo, Limit X, Pastor Helen Yawson, Sonnie Badu, Pastor Josh Laryea, the late Danny Nartey, Nii Okai, Soul Winners, Joe Mettle, Harmony Deh and many more.
Shofar has a goal to further encourage and celebrate the originality and meaningful expressiveness of jazz music.
News in Your Mail | Contact Us | Advertise With Graphic | Twitter
Generational Equity Advises Plumbing Master in its Investment by Newlook Capital – Business Wire
DALLAS–(BUSINESS WIRE)–Generational Equity, a leading mergers and acquisitions advisor for privately held businesses, is pleased to announce an investment by Newlook Capital, Inc. in its client, Plumbing Master, Inc. The investment closed June 2, 2022.
Located in Riverside, California, Plumbing Master (PM) is a commercial plumbing expert in sewer lines, sewer laterals, sewer connections, and trenchless technology. All of its technicians are experts in commercial plumbing, but not limited to: Installing water heaters and faucets, repairing and replacing toilets, water main lines, gas lines, slab leaks, and much more. PM generates a significant portion of its revenue from emergency repair with the remainder from scheduled maintenance. Plumbing Master’s owner, Andy Lozano, remains as an equity holder and will continue to lead PM’s nearly 100 employees in day-to-day operations
Newlook Capital (Newlook), located in Burlington, Ontario, Canada, is a private equity firm that currently manages two industry-specialized funds. Each fund has its own focused, experienced, and aligned management, and a defined strategy. Investing throughout Canada and the United States, with a preference for companies that exhibit a solid tangible assets base, Newlook is focused on, and current holdings include, industrial services in Canada and the United States, dental practices in Canada, and QSR in Canada.
Newlook seeks businesses that enjoy a distinct competitive advantage or have an established position in an industry that exhibits barriers to entry. Consolidation opportunities are of particular interest to them. Newlook targets acquisitions with earnings before interest, taxes, depreciation and amortization (EBITDA) between $1 million and $5 million.
“Andy has built an incredible business and we are extremely pleased to be partnering with him and further growing the company,” says Newlook’s CEO, Elroy Gust.
Gust added, “With this most recent partnership, we look forward to expanding the diversification of our revenue lines within our portfolio. With superior industry reputation and strong customer relationships, Plumbing Master is well positioned for growth.”
Andy Lozano, President of Plumbing Master stated, “I made a decision to transition the business in order to find an Investor with the resources, business expertise, and scale to manage and augment the rapid growth of the business. I found all of these in Newlook Capital and I look forward to the continued growth of Plumbing Master with their partnership.”
Generational Equity Executive Managing Director of M&A – Western Region, Stephen Crisham and his team, led by Sr. Managing Director – Western Region, Lori Galloway closed the transaction. Senior Managing Director, Ashok Tandon established the initial relationship with PM.
About Generational Equity
Generational Equity, Generational Capital Markets (member FINRA/SIPC), Generational Wealth Advisors, Generational Consulting Group, and DealForce are part of the Generational Group, which is headquartered in Dallas and is one of the leading M&A advisory firms in North America.
With more than 300 professionals located throughout 16 offices in North America, the companies help business owners release the wealth of their business by providing growth consulting, merger, acquisition, and wealth management services. Their six-step approach features strategic and tactical growth consulting, exit planning education, business valuation, value enhancement strategies, M&A transactional services, and wealth management.
The M&A Advisor named the company Investment Banking Firm of the Year three years in a row, Valuation Firm of the Year in 2020, and North American Investment Bank of the Year in 2022. For more information, visit https://www.genequityco.com/ or the Generational Equity press room.
Carl Doerksen
972-342-0968
[email protected]
Newlook Capital invests in Plumbing Master, a Generational Equity client.
Carl Doerksen
972-342-0968
[email protected]
Whitmer kidnap trial: Key FBI informant clashes with defense – Detroit Free Press
GRAND RAPIDS – “Big Dan,” a key witness and undercover FBI informant who infiltrated a group of militia members accused of plotting to kidnap Gov. Gretchen Whitmer, pushed back during cross-examination Tuesday against the notion he worked for the agency to boost his own profile.
“I wanted to protect lives,” Big Dan said. “I never asked for money. I never expected money. I wanted to prevent bad people from doing bad things.”
Big Dan, a 35-year-old postal contractor and U.S. Army veteran, gathered evidence for the FBI through secret audio recording devices and photographs at meetings and events held throughout the summer of 2020. His testimony is vital to the government’s case against Adam Fox and Barry Croft, whom prosecutors have described as the ringleaders in a plot to kidnap Whitmer in 2020.
Fox and Croft are being retried in federal court here after a jury earlier this year couldn’t reach a verdict in their case. Two codefendants were acquitted during that trial.
More:Informant ‘Big Dan’ returns in Whitmer kidnapping retrial as feds fight entrapment claim
More:Juror’s daughter-in-law got high with kidnap defendant Fox near Whitmer cottage
During cross-examination, defense attorneys noted that Dan was paid by the FBI — about $54,000 for his work from March-December 2020.
“You earned as much as you do hauling mail,” said Joshua Blanchard, one of the defense attorneys, during Big Dan’s cross-examination.
Big Dan fired back, saying he begrudgingly accepted to work undercover for the FBI after he became concerned that the militia group was plotting and training to kill police officers. The money he received was to reimburse expenses for attending events, buying meals and at one point moving to a new home after a militia member learned his previous address.
Blanchard continued to scrutinize Dan’s finances, showing the jury a receipt for a $3,399.99 Surface Book 3 laptop Dan purchased, along with a $199.99 docking station — both of which were reimbursed by the FBI.
Dan said the FBI agents he corresponded with, Special Agents Jayson Chambers and Henrik “Hank” Impola, had taken an interest in his personal life after he began working as an informant. It was the FBI who pushed for reimbursing the hardware, Dan testified, adding that he planned on using the laptop once he went back to college.
Cross-examination at times became combative, as Dan pushed against claims made by Blanchard, who is representing Croft in the case, and fellow defense attorney Christopher Gibbons, who is representing suspected plot leader Fox.
Gibbons highlighted text messages exchanged between Dan and his FBI handling agent, including one where the handler instructed Dan to “have (Fox) post his pictures” of Whitmer’s vacation home in Elk Rapids after Fox, Dan and one other individual visited the home in August 2020.
In another message, the FBI handler told Dan to “start talking details of the night op with Adam,” referring to the group scouting the vacation home the night of Sept. 12.
On Sept. 26, the handler texted Dan “We are good to start talking to Adam about the ruse for October 7.”
The ruse to which Chambers referred was a sting operation conducted when Dan led the group to a warehouse in Ypsilanti. The men thought they were meeting a contact for free tactical gear, like bulletproof vests, tactical belts and a Kevlar helmet, but were instead met by FBI agents.
Gibbons noted that the gear wasn’t a sign of any conspiracy, saying none of the men expected to obtain explosives at the warehouse. Gibbons also said that Dan was the one who organized the warehouse trip, not Fox.
More:Judge scolds defense in Whitmer kidnap trial: Stop wasting time talking ‘crap’
More:Meet the jurors in the Whitmer kidnap retrial — most don’t like the news
The attorney also highlighted Dan’s previous testimony that the group was supposed to meet at Buffalo Wild Wings for a meal after grabbing the fake gear. Dan had testified he had gone to the restaurant with others in the militia previously; Gibbons pointed out that Fox had been left out of those meetings to that point.
“Adam and I had been to the tavern in Luther,” Dan said.
“But not with the other guys, just you,” Gibbons replied. When Dan tried to explain why Fox hadn’t joined the group at Buffalo Wild Wings, Gibbons interjected.
“If you answered the questions I asked you to answer, we’d go a lot faster,” Gibbons said.
Defense attorneys have long argued Fox and Croft were entrapped by the FBI to create a plan to kidnap Whitmer. Both parties have said their clients are merely tough talkers who were frustrated by COVID-19 restrictions before they were pushed by overzealous FBI informants and undercover agents.
Fox, of Potterville, and Croft, of Bear, Delaware, were initially tried along with Daniel Harris and Brandon Caserta, but a jury was unable to deliver a verdict for them after finding Harris and Caserta not guilty. Two others, Ty Garbin and Kaleb Franks, previously pleaded guilty to their roles in the conspiracy.
Garbin was among the government witnesses to follow Dan on the stand Tuesday. Trial will resume Wednesday morning with his continued direct examination. Assistant U.S. Attorney Nils Kessler said Franks is expected to follow Garbin’s testimony.
Contact Arpan Lobo: [email protected]. Follow him on Twitter @arpanlobo.
Become a subscriber today.
21 Year Old Chesney Claire releasing NEW SINGLE August 26, 2022 – Daily Music Roll
(DailyMusicRoll Editorial):- Las Vegas, Nevada Aug 16, 2022 (Issuewire.com) – Las Vegas, NV– Well, it looks as though Chesney Claire has done it again with her new single, “The Problem”. Chesney Claire, a 21-year-old Pop/R&B award-winning singer/songwriter/producer, is all the rage this Summer as she keeps showing the world her talent(s).
Jumping from one red carpet event to the next is what’s happening now. In the last few weeks, Chesney has walked the red carpet and was a Presenter for “Best Set Design” at the LaJolla, CA Fashion & Film Festival. Just a few days later she was in Atlanta, GA presenting awards on the red carpet for the International Singer Songwriter’s Association.
Chesney has worked in every capacity from an award winner to Presenter of awards and has taken some bold strides in modeling. In August 2022 Chesney Claire made it through 7 rounds of the Maxim CoverGirl Contest and finished 4th in the world for the coveted “Maxim CoverGirl” front cover of the magazine but Chesney could not devote enough time to the contest- as she was traveling from state to state during that time. Nevertheless, Chesney was asked to do a model photoshoot for a very well-known Fashion Magazine in California while she was there at the International Film and Fashion Festival. The shoot drew lots of attention from the Fashion Industry and propelled Chesney Claire into the limelight in the Film Industry as well. Chesney is scheduled to star in a feature film with A-list actors that starts filming in March 2023 in Arizona.
Not only has she gained tons of experience in those fields, but her main passion is also music.
Last year Chesney won 25+ Music Industry Awards; including but not limited to Female Vocalist of The Year at the International Singer Songwriter’s Association, and at Indie Star Radio in Hollywood, CA. She won (4) World Songwriting Awards, a Nomination from The Hollywood Music In Media Awards, got her first original song in a Television reality series, earned a “Half A Million Streams” Spotify Award, and a coveted Josie Music Awards Nomination amongst others.
Now Chesney Claire is learning music production and engineering.
Chesney won a Hermes Award this year for her music video that accompanied her latest single release “Not In The Mood“. That song was written by Chesney, produced by Chesney, and engineered by her.
Chesney’s latest release “The Problem” is scheduled to debut on August 26th, 2022- And it’s already been nominated to receive a Grammy at this year’s Grammy Awards. (Not to mention that Chesney just also became a voting member of the Recording Academy- Class of 2022). Chesney’s nominations are for “Best New Artist” and “Single of The Year” and although Academy Members may not know Chesney yet, her talent and absolute resolve to make it in the Music business shines brighter than all the stars in the sky.
“I know God has blessed me with undeniable talent. My goal is to make Him proud by using my talents to the best of my ability. I want to shine! I have a lot of Angels to make proud,” Chesney says.
Quite sure there will be many more opportunities for Chesney Claire as she continues down this road to international stardom.
“I’m now writing all my own songs. I’m producing my own songs, and I am engineering everything myself. Hopefully, that will count for something and help advance my career quickly“, she said.
The new song “The Problem” is a saucy, hot tropical Afro Pop dance vibe. It will be available on all streaming platforms on August 26th so please stream it. Listen on repeat, and share.
Chesney has just been commissioned to write a theme song for another upcoming feature film and is working on that now.
We all see great things in her future. She’s the hardest-working Female in the entertainment industry right now and if you haven’t heard of her, it’s time to do some research!
You can Google “Chesney Claire” to learn more. You can visit her website: www.chesneyclaire.com. You can contact her manager(s), David Davidian (Tour Manager for Alice Cooper/The Hollywood Vampires) at [email protected] or her Music and Television sync manager, Charlie Mac at [email protected]
Stay tuned. There’s so much more to come!
http://www.chesneyclaire.com
This article was originally published by IssueWire. Read the original article here.
Copyright © 2022 DailyMusicRoll.
Why is 'outside-zone' offense so hard to master? Ed McCaffrey explains – NBC Sports Boston
Watch
The storyline dominating Patriots training camp so far is the implementation of a new offense. The post-Brady/McDaniels switch to a scheme that’s going to be more user-friendly has been a slow and hard-to-watch slog on many days.
This is not an entire “out with the old, in with the new” undertaking, sources have told me. Elements of the offense the team’s used for years will remain. But there’s a pivot in the run game to using more stretch or “outside zone” concepts that, ideally, will become the basis for incorporating new pass-game concepts.
Don’t expect the Patriots to just throw up their hands because it’s looked ugly. Mac Jones‘ words last week were all the evidence you need to see that’s not happening.
“I’m going to figure it out,” Jones said, via Mike Reiss of ESPN.com. “I always have. I always will. At the end of the day, you’re going to have your ups and downs with anything new, but I’ve learned a lot of different systems and the guys around me have too. We know what football looks like, we know what a good play looks like. … It just needs to be more consistent. We all trust in each other at the end of the day. When there’s 10 people that look into my eyes, I know they’re going to trust me to do the right thing on game day.”
Why’s it been so hard to master and why’s it looked so rough in practice? Because the outside-zone scheme is complex.
Last Thursday, I asked former Denver Broncos great Ed McCaffrey, who came up in the Mike Shanahan offense that beautifully employed outside zone (and sometimes took Bill Belichick’s defenses to the woodshed) what a switch to outside zone would entail.
What are you gonna be good at? You can’t be great at everything. You just can’t. So what are you gonna hang your hat on? And if it’s gonna be outside-zone it takes time.
McCaffrey, who was co-hosting on SiriusXM NFL Radio when I asked him, went into a detailed explanation that really lays out the intricacies.
“Some offensive line coaches are insulted when you call it outside zone,” McCaffrey began. “They like to call it wide zone and they call inside zone ‘tight zone.’ Rarely when you are running an outside zone play does it go outside. It usually cuts back somewhere in between the edge defender and the hash mark. Some people call it stretch, doesn’t matter what you call it, it is a whole system.
“But just the outside run play itself is an entire system involving multiple two-man combinations on the front side and back side of the play. It gets more complicated when you introduce different formations and different personnel groupings. So, for instance, you could run a wide-zone play out of 21 personnel (two backs, one tight end) which is what most pro-style West Coast offensive coaches consider their base offense. That involves a fullback and a tight end.
“(The complexity rises when you get into) are you running a full-flow to the tight end with the fullback? Are you running split-flow to the tight end with the fullback going the opposite way? Are you running to the open side with the fullback?
“In addition to that, every one play is actually two because you’re coaching the front side and the back side. Center and guard have to communicate, guard and tackle have to communicate, tackle and tight end have to communicate. And then there’s 36 minimum basic fronts that involve different two-man combination calls. So if defenders ‘mug and bug’ bringing linebackers up you can potentially turn that into a three or four-man combination play. Communication on the line just for that — we aren’t even discussing 11 personnel or 10 personnel motions and shifts — is a whole entire language.”
Patriots Talk: What’s the end goal of the Patriots playing ‘process’? | Listen & Subscribe | Watch on YouTube
McCaffrey mentions the fullback. The Patriots made an interesting move with that position in the offseason, saying goodbye to Jakob Johnson and letting him know the straight-ahead battering ram style was going away from their offense.
So the outside-zone style McCaffrey’s referencing from 21 personnel may not be the Patriots’ base formation. Either way, the concepts are still complicated.
“If you haven’t done it then it does take time just to install the outside zone scheme,” McCaffrey explained. “Just one play is really like 20 plays (because of the variations in personnel and options). So to put in an outside zone scheme with multiple personnel groupings … think of the time it takes to put that in.
“In the NFL you have time to put in multiple blocking schemes and all teams do. But how much time are you going to spend doing that? Time is our most valuable asset. So to get guys up to speed and be really good at the outside zone schemes which involve multiple combination blocks versus various fronts, to get really, really good at that takes time.
You can’t be great at everything but as a playcaller you still want enough variation to keep defenses on their heels and attack certain fronts. But the big decision is, ‘How much time am I gonna spend on this other stuff? What am I gonna live out of?’ It takes time to learn. A lot of time. And really repetition. Practicing their fits on those two-man combinations and communicating.”
McCaffrey, whose son Christian is a standout with the Panthers, currently coaches at the University of Northern Colorado. He’s got an appreciation for what the Patriots staff is wrestling with.
“Think of wide zone as a whole language in and of itself,” he said. “And with that, you only have so much time in a day to get really good at something. And that’s the question for head coaches and coordinators. ‘What do we want to be great at?’ If you want to be great at outside zone, you gotta spend a lot of time running outside zone.
“Then you have to pair that with all of your play-action passes. Your play-action passes are gonna work off of the runs you run well and have success running. … and with play-action it all comes off of the run game. What are you gonna be good at? You can’t be great at everything. You just can’t. So what are you gonna hang your hat on? And if it’s gonna be outside-zone it takes time.”
And it’s going to take time for it to look good. Which is what we’re seeing in Foxboro currently.
1 dead, 1 injured in shooting outside Atlanta recording studio – WSB Atlanta
1 dead in shooting outside Atlanta recording studio
ATLANTA — Police are investigating a shooting that left one person dead and another injured in southwest Atlanta.
Channel 2′s Darryn Moore is at the investigation location off 2945 Stone Hogan Connector where there is a large police presence.
The address belongs to a building that holds several recording studios. It is unknown if the victims were connected to any of the businesses.
TRENDING STORIES
The shooting happened around 3:30 a.m. Investigators told Moore that it appears the shooter fired from the driveway of the business.
Moore counted at least 26 evidence markers placed where shell casings were found.
Atlanta homicide detectives said a man in his late 20s died at the scene. The second victim is stable at Grady Memorial Hospital.
No description of the shooter is available at this time. The victims’ names have not been released.
[DOWNLOAD: Free WSB-TV News app for alerts as news breaks]
IN OTHER NEWS
DeKalb County residents say they are ‘powerless’ and ‘frustrated’ by crime in their community Neighbors in DeKalb County told Channel 2 Action News they feel ‘powerless’ and ‘frustrated’ by criminal activity they see happening in their community.
©2022 Cox Media Group
© 2022 Cox Media Group. This station is part of Cox Media Group Television. Learn about careers at Cox Media Group. By using this website, you accept the terms of our Visitor Agreement and Privacy Policy, and understand your options regarding Ad Choices. Manage Cookie Preferences | Do Not Sell My Information
Why I chose to get my master's degree in nursing online – Fortune
Katarina Maric had logged a decade as a registered nurse before tackling her master’s degree in nursing. In fall 2020, Maric enrolled in Marymount University’s online master’s program in nursing to become a family nurse practitioner. “I always knew I wanted to be a nurse and continue my education,” she says. “I wasn’t sure of the right time.”
Taking courses online worked best for Maric—she’s busy at home in a suburb of Washington, D.C., with her 3- and 5-year-old boys. Selecting the best time to begin and the program was a process, she says.
Turns out, Maric’s decision process is fairly typical, says Alison Colbert, a professor at Duquesne University’s School of Nursing. “I always joke, no one says, ‘Well, I wish I waited longer to get my graduate degree,’” says Colbert. “But an equal number of people say, ‘I needed the experience and the perspective. I had my life.’”
Because nursing is predominantly a female profession, that inevitably informs how schools approach education—particularly for advanced degree programs, Colbert says. Schools offering advanced nursing degrees need to be flexible to capture students who need to work, raise children, care for parents—or simply find the time to save money, arrange for childcare, or shift work schedules to make room for school to begin with.
For Maric, who will graduate in December, navigating the decision to enroll in Marymount’s online MSN program took time and thought. She sat down with Fortune to share her experience and offer some advice for prospective students considering a similar path.
The following interview has been edited for brevity and clarity.
Fortune: Why did you go back to school?
Maric: Right before I went to Marymount, I got sick and was navigating the health care system and got really frustrated. I decided that as soon as I felt better, I was going to go back.
How does the master’s degree line up with your professional goals?
It’s an advanced practice nursing degree. I will be able to function as a primary care provider. As an RN, you follow orders, and there’s not much autonomy. As a family nurse practitioner, I can work anywhere—in family practice or in a specialty with patients of all ages. There are different routes; they say it’s like cradle to grave, from newborns to end of life. I can work with anybody and function as their primary care provider and work in any specialty.
I wanted to expand my professional knowledge and felt like it was a natural progression to move into a more advanced role, more of a health care provider role. As a school nurse, I worked at a middle school that had a lot of underprivileged students. I was realizing I was often the only health care professional they’d see. Their parents work hourly jobs and can’t afford to take off to take them to the doctor.
At first, I thought it would be cool to work in the school system as a nurse practitioner in a clinic in the school system. That was my initial thought: Maybe I could bridge that gap. But I’ve since learned that in Virginia, in order to practice independently as a nurse practitioner, you have to have a practice agreement with a physician for five years before you can apply to practice independently. Maybe down the line somewhere; I won’t be able to do that right out of school.
How did you choose your program?
I did a ton of research. Marymount is semi-local, as Arlington is 40 minutes from my house. I wanted support. As soon as I talked to a student success coach and adviser, I got the feeling they have a good support system.
Even though the program is mostly online, I wanted to be able to go to campus if I had to. They have student success coaches that stay with you throughout the program. It’s my first experience with online education. I want to be present for my kids, and the traffic in D.C. is crazy. I didn’t want to have to drive and waste time driving, when I could be spending that time studying.
Are you working while going to school?
I decided to stop working so I can spend time with my kids and go to school. My kids are only in school for four hours a day, so it doesn’t leave too much time. As part of the program, we do work. I work three days a week right now in an internal medicine practice as part of the program. Marymount finds clinical placement for you and they set you up. They found this practice for me.
We also have a nurse residency program. For that, we have to go on campus for one weekend—when we are done with our core classes, before we start clinical rotations. We see patients on campus and do physical exams on them to show our skills.
Has your decade of experience as an RN helped with that piece of it?
I think it really helps that I worked for 10 years—understanding the health care system, understanding patients and their stories, talking to people. There is an art in talking to patients on sensitive subjects, especially if they have serious trauma or something they are embarrassed about. It helps to have that nursing background so you know how to approach patients. We are building on knowledge we already have, from ecology to pathophysiology.
What has surprised you most about your experience?
The main surprising thing is related to it being online. I thought you just do it and get your work done week by week. I’m surprised by the relationships I’ve built with people—you don’t feel like you’re online, you have support all of the time. We have [mandatory] synchronized sessions over Zoom every other week. We are all there and talking and interacting. I’ve become great friends with my classmates.
Our professors have gone above and beyond, taking out extra time to meet with us on Zoom or help us find clinical placement. For people that don’t have an established network—they pulled out their own contacts and helped set them up. The professors are extremely involved and truly care about our success. They want us to have the tools we can to perform well. I thought it might be lonely to do it online, but it’s definitely not.
See how the schools you’re considering fared in Fortune’s rankings of the best master’s degree programs in nursing, computer science, cybersecurity, psychology, public health, business analytics, and data science, as well as the best doctorate in education, and part-time, executive, full-time, and online MBA programs.
Glowing snailfish riddled with antifreeze protein discovered off Greenland coast – Livescience.com
Live Science is supported by its audience. When you purchase through links on our site, we may earn an affiliate commission. Here’s why you can trust us.
By published
This helps protect the tadpole-like fish from the cold.
Scientists drilling deep into an iceberg off Greenland have discovered a fish with glowing green antifreeze coursing through its veins.
The juvenile variegated snailfish (Liparis gibbus) contained the “highest expression levels” of antifreeze proteins ever reported, a new study found.
Similar to how antifreeze helps to regulate the temperature of a car’s engine in extreme conditions, certain species have evolved to have similar protection, especially those living in frigid habitats such as the polar waters off Greenland.
“Antifreeze proteins stick to the surface of smaller ice crystals and slow or prevent them from growing into larger, and more dangerous, crystals,” study co-author David Gruber, a research associate at the American Museum of Natural History (AMNH) and a distinguished biology professor at City University of New York’s Baruch College, told Live Science in an email. “Fish from both the North and South Poles independently evolved these proteins.”
Related: Robotic sub will explore underside of Greenland’s glaciers in a first
Antifreeze proteins were first discovered in some Antarctic fish nearly 50 years ago, according to the National Science Foundation (opens in new tab).
Unlike certain cold-blooded species of reptiles and insects, fish are unable to survive when their bodily fluids freeze, which can cause grains of ice to form inside their cells and essentially turns them into fish Popsicles.
“The fact that these different antifreeze proteins have evolved independently in a number of different — and not closely related — fish lineages show[s] how critical they are to the survival of these organisms in these extreme habitats,” John Sparks, a curator in the AMNH’s Department of Ichthyology and co-author of the study, told Live Science in an email.
Snailfish produce antifreeze proteins “like any other protein and then excrete them into their bloodstream,” Gruber said. However, snailfish appear to be “making antifreeze proteins in the top 1% of all other fish genes.”
“The snailfish was one of the few species of fish living among the icebergs, in the crevices,” Gruber said. “It was surprising that such a tiny fish could live in such an extremely cold environment without freezing.”
It’s also rare for Arctic fishes to exhibit biofluorescence, which is the ability to convert blue light into green, red or yellow light, since there are prolonged periods of darkness, especially in the winter, at the poles. Normally this characteristic is found in fish swimming in warmer waters. This is the first reported case of an Arctic fish species exhibiting this adaptation, according to an AMNH post (opens in new tab).
The scientists further examined the biofluorescent properties of the snailfish and found “two different types of gene families encoding for antifreeze proteins,” according to a separate statement, an adaptation that essentially helps them avoid turning into frozen fish sticks.
—Bizarre translucent ‘Jell-O fish’ pulled up from icy depths in Alaska
—What’s the weirdest sea creature ever discovered?
—Strange ‘alien’ holes discovered on the ocean floor
This mind-boggling level of antifreeze production could help this species adapt to a subzero environment, according to the statement. It also raises a question about how snailfish will fare as ocean temperatures increase as a result of global warming.
“Due to rapidly warming waters in the Arctic, these cold-water-adapted species will also have to compete with warmer-water species that are now able to migrate north and survive at higher latitudes (and that won’t need to produce metabolically costly antifreeze proteins to survive in the warmer Arctic waters),” Sparks said. “In the future, [antifreeze] proteins may no longer provide an advantage.”
The findings were published Aug. 16 in the journal Evolutionary Bioinformatics (opens in new tab).
Originally published on Live Science.
Jennifer Nalewicki is a Salt Lake City-based journalist whose work has been featured in The New York Times, Smithsonian Magazine, Scientific American, Popular Mechanics and more. She covers several science topics from planet Earth to paleontology and archaeology to health and culture. Prior to freelancing, Jennifer was a reporter at Interior Design Magazine, and before that she held an Editor role at Time Inc. Jennifer has a bachelor’s degree in Journalism from The University of Texas at Austin.
Stay up to date on the latest science news by signing up for our Essentials newsletter.
Thank you for signing up to Live Science. You will receive a verification email shortly.
There was a problem. Please refresh the page and try again.
Live Science is part of Future US Inc, an international media group and leading digital publisher. Visit our corporate site (opens in new tab).
© Future US, Inc. Full 7th Floor, 130 West 42nd Street, New York, NY 10036.
Looking to invest in transcription services? The most important factors for legal industry professionals to consider. – JD Supra

[author: Sandra Redavid]
While there’s been a rising demand for audio and video transcription services across all industries, the legal transcription market accounts for nearly 30% of the transcription market as a whole. And demand for legal transcription in the U.S. is expected to account for 90% market share in North America through 2029. These statistics align with my experience – compared to pre-COVID, since 2020, I’ve seen a 60%-70% uptick in audio and video transcription requests.
What’s behind this increased demand? The pandemic was, and is, a definite factor. The abrupt shift from in-person to virtual everything meant people in all industries needed new ways of getting transcriptions made. The rise of digital platforms like Zoom, Microsoft Teams, and Google Meet also played a role, as more professional interactions than ever were recorded as video and audio files. Transcripts can help for meetings where you just want the notes recorded or want to remember something important that wasn’t written down.
When it comes to record keeping requirements, in some industries, having just an audio recording or video recording is enough. In others, like the legal industry, a hard-copy transcript is required because an accurate record of a legal proceeding is always necessary. No matter the reason, as audio and video transcription demands ramp up, legal organizations want to pay careful attention to how and where recordings are transcribed, as this could drastically impact legal outcomes.
In terms of the type of recording, there is no rule in the legal industry that specifies you must use video recordings over audio recordings or vice versa. Both recording formats can be filed in court. However, you are not permitted to file only an audio and/or video recording—a typed transcript must accompany the filing. Whether a recording of a wiretap, a 911 call, a police interview, a deposition, or a Senate hearing, any digital file must include a written transcript.
As a transcription services provider, U.S. Legal Support meets both HIPAA and HITECH compliance requirements and takes privacy and confidentiality extremely seriously. This means we never sell any copies of transcripts, nor share transcripts with any party other than the original submitted. We also take extra care that our transcripts are sent via a secure, non-shareable link within our SOC 2 Type 2 and HIPAA compliant client portal. In addition, all transcripts are kept in a secure database that follows our SOC 2 Type 2 compliant protocols as well.
We take security another step further by ensuring our “people” are secure as well. Transcribers are required to sign business agreements, including non-disclosure agreements. You can learn more about how U.S. Legal Support wraps all of our services in industry-leading cybersecurity efforts here.
There are so many ways to create recordings these days, but when it comes to files needed for audio and video transcription, some formats are definitively better than others. If you’re planning to have something transcribed for a legal or court proceeding, it is important to think through the following.
#1: The platform
In terms of virtual meeting platforms, I’ve found that Zoom is one of the better ones for translating an audio or video recording into a transcript. It provides you with speaker identification and displays who is speaking at a specific time within the recording.
Other platforms, like Google Meet, won’t tell you who is speaking. Unless it was captured on video and you are supplied with a name and what the speaker was wearing, you can have a hard time identifying speakers.
#2: The mic
The quality of a recording really comes down to the audio and where the mic is placed. If the person talking is across the room and not close to the mic, like on a speaker phone, it’s much harder to pick up his or her voice and there’s a better chance of misheard or dropped words.
The quality of the connection itself also plays a big role. This can come into play with wire taps or cell phone recordings of interviews, or if the person recording a conversation has a poor Wi-Fi connection.
#3: The speakers
Accents and dialects can also impact transcription accuracy. For this reason, when dealing with recordings where international speakers are involved, it’s vital to use a real person to create the transcript—and if it’s someone who is intimately familiar with the language and dialect, and can transcribe it with great accuracy, all the better.
At U.S. Legal Support, our team has translators who are fluent in multiple languages and dialects. When we work with a legal entity on transcribing an audio or video recording where dialects or accents are a factor, we’re able to assign that recording to someone who understands both the phrases and words commonly used in that dialect, as well as different legal jargon that might be used within the speaker’s court system.
#4: The terminology
Using a litigation support resource who understands the area of expertise covered by the recording also helps in the process of cleaning up the audio prior to transcription. If it’s a recording of a financial board meeting, for example, we assign someone who understands the financial industry and its terminology. This ensures a more customized, accurate experience and transcript.
Given the nuances in transcript requirements, here are considerations for legal professionals when looking to invest in audio or video transcription services.
What to avoid:
Looking ahead
I believe audio and video transcription holds big promise for the legal industry in the coming years. I also believe the further development and refinement of additional technologies is going to greatly enhance the transcription world, too.
See more »
Refine your interests »
This website uses cookies to improve user experience, track anonymous site usage, store authorization tokens and permit sharing on social media networks. By continuing to browse this website you accept the use of cookies. Click here to read more about how we use cookies.
Back to Top
Explore 2022 Readers’ Choice Awards
Copyright © JD Supra, LLC
Leaked audio reveals Liz Truss said British workers needed ‘more graft’ – The Guardian
Exclusive: Tory leadership frontrunner suggested Britons lacked ‘skill and application’, in echo of ‘idlers’ row
Liz Truss, now the Tory leadership frontrunner, launched an astonishing broadside against British workers, saying they needed “more graft” and suggesting they lacked the “skill and application” of foreign rivals, the Guardian can reveal.
In a leaked recording, the then No 2 at the Treasury also risked pitting Londoners against the rest of the country by attempting to explain the difference between the capital and other regions in the UK.
Truss, who has put patriotism at the heart of her leadership campaign, suggested the disparity was “partly a mindset or attitude thing”.
The comments were made when Truss was the chief secretary to the Treasury, a post she held until 2019. In the recording she intimated that there seemed little desire to change the working culture so that the UK could become more prosperous.
The highly disparaging remarks echo a controversial passage about British workers being among the “worst idlers in the world” in the book Britannia Unchained, which she co-authored in 2012 when she was a new backbench MP seeking to make her mark as a neo-Thatcherite.
In the first televised head-to-head Tory leadership debate last month, Truss claimed she had not written the offending chapter and blamed her fellow author Dominic Raab instead.
She told the BBC presenter: “Each author wrote a different chapter. Dominic Raab wrote that chapter – he’s backing Rishi Sunak.”
Raab later claimed that the authors, who also included Priti Patel and Kwasi Kwarteng, had taken “collective responsibility” for the book, adding: “It’s up to Liz to explain why she’s changed her view.”
In the leaked recording, Truss claimed that the book had been “mischaracterised” at the time of its release a decade ago, but gave no detail as to how she felt the passage had been misrepresented.
Truss’s remarks about the productivity of workers outside London could be particularly damaging as earlier this month she was forced to make a U-turn on plans to cut civil service pay outside the capital after a furious outcry from Conservative MPs.
She claimed there had been a “wilful misrepresentation” of her policy – despite her campaign having published specific details – but confirmed she was abandoning plans for regional pay boards for civil servants or public sector workers.
Sign up to First Edition
Archie Bland and Nimo Omer take you through the top stories and what they mean, free every weekday morning
London had the highest productivity level of any UK region in 2020, with output per hour more than 50% higher than the median, according to the Office for National Statistics.
However, this is widely believed to be the result of large multinationals being based in the capital, higher engagement with research and development, the size of firms and the level of exports, and the transport infrastructure.
In the leaked recording, Truss began: “I once wrote a book about this which got mischaracterised – British workers produce less per hour than … and that’s a combination of kind of skill and application.”
She went on: “If you look at productivity, it’s very, very different in London from the rest of the country. But basically … this has been a historical fact for decades. Essentially it’s partly a mindset and attitude thing, I think. It’s working culture, basically. If you go to China it’s quite different, I can assure you.”
The minister, who had close oversight of public spending, added: “There’s a fundamental issue of British working culture. Essentially, if we’re going to be a richer country and a more prosperous country, that needs to change. But I don’t think people are that keen to change that.
“There’s a slight thing in Britain about wanting the easy answers. That’s my reflection on the election and what’s gone before it, and the referendum – we say it’s all Europe that’s causing these huge problems … it’s all these migrants causing these problems. But actually what needs to happen is more … more graft. It’s not a popular message.”
Truss was questioned about the Guardian’s revelations during a Tory leadership hustings in Perth, and appeared to confirm she still believed British workers were not as productive as they should be.
Asked by Colin Mackay, STV’s political editor and chair of the hustings, whether she stood by those remarks, Truss said: “I don’t know what you’re quoting there [but] what we need in this country is more productivity and we need more economic growth.
“The thing is we don’t have enough of is capital investment, which why it’s important to get more investment in the whisky industry and North Sea.”
The shadow work and pensions secretary, Jonathan Ashworth, said: “With wages shrinking thanks to Tory failure to bring inflation under control and years of lacklustre growth, it’s grossly offensive for Liz Truss to effectively brand British workers lazy.
“I would have hoped she had moved on from the days of her Britannia Unchained fiasco, but it seems that is the blueprint for her prospective government. Workers across the country are working all hours to keep a roof over their heads, put food on the table and provide for their families. Liz Truss should be helping working people to cope with this cost of living crisis, as Labour this week outlined we would do, not peddling this offensive nonsense.”
60 Years of Stereophile! – Stereophile Magazine

“Okay, if no one else will publish a magazine that calls the shots as it sees them, I’ll do it myself,” he later recalled thinking, adding, “I must have been out of my mind!”
Six decades later, Paul McCartney and Ringo Starr are still touring (though not together), and the magazine JGH founded, now just called Stereophile, is celebrating 60 years of continuous publication (footnote 1).
This longevity is a rarity in magazine publishing. When that first issue appeared in readers’ mailboxes, there were three dominant audio magazines in the US: High Fidelity, Audio, and Stereo Review (then called HiFi/Stereo Review). High Fidelity was published from April 1951 until July 1989, when it was bought by Stereo Review‘s then-publisher and shuttered. Stereo Review was first published in 1958, with the title HiFi and Music Review. In 1999, it merged with Video magazine. In 2000, it was retitled Stereo Review’s Sound & Vision—then later just Sound & Vision—to emphasize its change from an audio magazine to one covering home theater as well.
Audio started life in 1947 as Audio Engineering, then dropped the word “engineering” in 1954. Audio‘s final issue was a combined February/March 2000 issue, after which it merged with Sound & Vision. In a strange twist, Stereophile‘s then-publisher bought Sound & Vision in 2013; Sound & Vision is now owned, like Stereophile, by AVTech Media, which also publishes the only English-language audio magazine that has exceeded Stereophile‘s 60-year (so far) lifespan: the UK’s Hi-Fi News & Record Review, which published its first issue in June 1956.
What could lie behind this magazine’s longevity? In an article I contributed to Hi-Fi News & Record Review‘s 50th anniversary issue, in June 2006, I wrote: “Monthly magazines, with their focus on the new, the exciting, the cool, are inherently ephemeral things. Some readers react to this reality by merely glancing at a few features as the magazine wends its way from mailbox to recycling bin. However, other readers pore over each issue’s content, using it to stay abreast of what is happening, to feed their passion for the subject. Such readers are drawn to a publication that shares their passion, that doesn’t talk down to them, that is unfailingly honest with them, that informs and educates as it entertains, that gives them more than they expect with respect to both breadth and depth of content.”
These words have applied to Stereophile since its founding, coupled with J. Gordon Holt’s vision that the optimal way to judge audio components was to do what end users did: listen to them! “Dammit,” Gordon said, “if nobody else will report what an audio component sounds like, I’ll do it myself!”
The audio space was simpler in the 1960s. In its first 12 issues, published as Vol.1 from September 1962 through May 1966, Stereophile reviewed 84 products from 36 manufacturers, doing a relatively complete job of covering what was available then. By contrast, the magazine’s 12 issues published in 2021 (Vol.44) contained reviews of 125 products from 101 manufacturers and couldn’t come close to being comprehensive. Even so, other than stablemate Hi-Fi News, there isn’t an audio magazine that can rival the depth of Stereophile‘s audio product reviews.
Of the 36 brands reviewed in Vol.1, several have been in continuous existence since: Bang & Olufsen, Beyerdynamic, Grado, Koss, Marantz, McIntosh, Nagra, Neumann, Ortofon, and Shure. The reviews in 2021 included products from Marantz and McIntosh, as well as some from other brands that have been in continuous existence since Stereophile‘s first issue was published but weren’t reviewed in Vol.1: KEF, Klipsch, SME, and Sony.
J. Gordon Holt, who passed away in 2009, created not just a magazine but a community of what came to be called audiophiles. Stereophile‘s team of contributors has played an important role in that community. There have been too many to list here, but I will mention those who are no longer with us and who played major roles in creating the magazine’s identity through those 60 years: Art Dudley, Margaret Graham (Mrs. Holt), Peter W. Mitchell, Wes Phillips, Robert J. Reina, Rick Rosen, John Swenson, and Steven W. Watkinson, as well as art director Daniel Bish, managing editor Debbie Starr, and advertising representative Ken Nelson.
Another writer who also played a major role at the magazine for more than a quarter-century, but who is very much alive, is analog guru Michael Fremer. His first Analog Corner column appeared in July 1995 (footnote 2), and in 2012, he was hired as a fulltime staff member, not only to continue contributing Analog Corner and equipment reviews but also to edit AnalogPlanet.com, Stereophile‘s sister website. Sadly, Mikey (as he likes to be called) resigned from both Stereophile and AnalogPlanet.com in June. His final column appears in this issue, along with his final review, of the humongous Gryphon Apex Stereo amplifier.
All of us at Stereophile wish Mikey well in his new ventures.
Footnote 1: You can find a timeline of Stereophile’s first 40 years here and articles celebrating the magazine’s 500th issue here and here, and a related photo gallery here.
Footnote 2: Analog Corners through November 2005 are archived here; from October 2017 onward they can be found here, with earlier columns added each month.
- Log in or register to post comments
You (JA2) can ban me if you want, I really don’t care at this point.
I’m sure you guys can see what’s happening, and I really get that you have a limited editorial budget to work with, but in any case, isn’t there anything you can do to make this site more interesting for the readers?
Right now, AnalogPlanet is fresher than you guys. Think about that. Personally, I’m not interested in record reviews, so they’re not an alternative for me.
Hi
To YOU, maybe. Not to many many loyal readers out there, including yours truly.
“Personally, I’m not interested in record reviews”
Who cares !
Jack L
PS: I’m still awaiting yr reply to my question to you: “Are you really an EE ??”. I’m all ears.
For you to answer his question as well.
Hi
Talk is cheap! I want to read his posts in Audiophile forums, if any at all, in electrical engineering language.
Back to the same topic on $4,000+ interconnects, I want him to show me his EE knowledge in audio cables: the formula to calculate the “characteristic impedance” of an audio cable.
Otherwise please shut up & back off.
Jack L
Hi MatthewT
I meant to tell the guy who challenged me with such redundant personal question to “shut up & back up” if he could not come with the formula to calculate the “characteristic impedance” of an electrical cable.
FYI, audio electronics is only my leisure hobby to supplement my classical music enjoyment. It’s kinda ‘peanut’ vs my EE experience.
My professional EE experience is much more subtantial:
providing engineering solutions to the power industries for deacades, e.g. to power companies for their HV power distribution.
I dealt with power company engineers day in day out.
Jack L
I wouldn’t say that Stereophile is dying. In fact, of all the sites that I visit, Stereophile is first on my list. Sure, I visit Analogue planet, and stop by Twittering and maybe even Darko. But, my tastes were formed by the late Art Dudley. This is the first place I stop by to look for reviews of equipment that I am considering to purchase. Perhaps, I’m more in the maker end of the spectrum.
So, maybe a return to the idea of being a maker audiophile is in order. I’m certain my grandfather purchased your fist issues, because he had a subscription when he was 90 decades ago. He also designed audio equipment for RCA and such. So, I’m in the market for a SUT for an EMT TSD. I also just found out that EMT is making new tonearms! Yay!!! My next set of speakers I’ll be building with Rulliet drivers, but hey, I do well so a set of KEF LS60s might show up.
All in all, death looks somewhat premature.
I remain a steadfast fan.
We are definitely in the time of diminishing ‘hard copy’ circulation, but that is certainly not unique to Stereophile.
Plus, evil empires now rule the landscape, so I am even more impressed that Stereophile can maintain some degree of the appearance of autonomy.
Plus, the hobby is so different now.
In 1962, a person earning the median wage could participate in the hobby at the highest level. Now, it seems the average demo at a given hi fi show is double the median price for an entire home. That part might be a factor in how we are seen by non-audiophiles.
*Don’t get me wrong, I understand the great affordability of entry and middle ground gear. It’s just what JA1 called that price event horizon.
I remain a steadfast fan.
Thanks.
…the hobby is so different now. In 1962, a person earning the median wage could participate in the hobby at the highest level. Now, it seems the average demo at a given hi fi show is double the median price for an entire home. That part might be a factor in how we are seen by non-audiophiles.
*Don’t get me wrong, I understand the great affordability of entry and middle ground gear. It’s just what JA1 called that price event horizon.
I used that term in my January 2017 As We See It: www.stereophile.com/content/price-event-horizon. I was referring to an an earlier essay – www.stereophile.com/content/upward-price-spiral – in which I wrote “if an audio manufacturer has to gross a certain amount of revenue each quarter to meet payroll, cover fixed expenses, invest in parts for the next quarter’s production, and pay the interest on any capital they’ve borrowed, the least risky business strategy is to bring to market a very small number of very expensive products.” But that strategy make so many great products financially out of reach for regular people, like you and me, Anton.
John Atkinson
Technical Editor, Stereophile
I’m not complaining, but the Entry Level column needs to make a comeback.
Hi
Just last nite I streamed “Concert for George (Harrison)” performed 11/29/2002 at the Royal Albert Hall on my 4KUHD TV with sound feeding my audio rig.
What a full-house musical event! Paul McCartney & Ringo Starr were seen playing together with George’s then young son: Dhani, who looked so much like his dad !
Jack L
Hi
Though JGH outlived many audio celebrities back then (He was 79 !!), he could have lived much longer should he give up his addicted smoking. He only stopped smoking 10 years after he got the painful throat cancer !
This reflected his determination of running a readers-friendly audio magazine aginst those well established “showcases for advertisers”.
Jack L
A Salutation to the original JGH! A 2nd salutation to J.A. for his hard work and passion at the helm of Stereophile. May J.A.(2) have the same flair and zest for our beloved hobby. I have been a dedicated subscriber sine 1993 to present date.
Think about adding a “Dealers Showcase” several times per year to recognize those brick-and-mortar retailers still in the field.
Happy Listening!
J.A.(3)
60 years is an utterly amazing accomplishment! John’s work, Jim’s work, Larry’s… and of course Gordon, there has been no other magazine that provides the insight, knowledge and truth that Sterophile has provided over the past 6 decades!
I would even argue that they are the best, most interesting and best written magazine of ALL TIME.
To those that would say much less or worse about these fine writers and editors… you have no place to disparage with these professionals!
Thank you truly, for the most thought provoking periodical extant.
Wow! Remembering the 25th anniversary makes me feel old!
Congratulations to all. My 35 years with the magazine has provided tons of enjoyment, especially early on and for years when I read it ravenously in less than a day, then began the wait for the next issue (pre-internet!).
Bill
…to the team!! Pls keep up the great work!!
best regards
HB
You (JA2) can ban me if you want, I really don’t care at this point.
I’m sure you guys can see what’s happening, and I really get that you have a limited editorial budget to work with, but in any case, isn’t there anything you can do to make this site more interesting for the readers?
Right now, AnalogPlanet is fresher than you guys. Think about that. Personally, I’m not interested in record reviews, so they’re not an alternative for me.
Hi
To YOU, maybe. Not to many many loyal readers out there, including yours truly.
“Personally, I’m not interested in record reviews”
Who cares !
Jack L
PS: I’m still awaiting yr reply to my question to you: “Are you really an EE ??”. I’m all ears.
For you to answer his question as well.
Hi
Talk is cheap! I want to read his posts in Audiophile forums, if any at all, in electrical engineering language.
Back to the same topic on $4,000+ interconnects, I want him to show me his EE knowledge in audio cables: the formula to calculate the “characteristic impedance” of an audio cable.
Otherwise please shut up & back off.
Jack L
No.
Hi MatthewT
I meant to tell the guy who challenged me with such redundant personal question to “shut up & back up” if he could not come with the formula to calculate the “characteristic impedance” of an electrical cable.
FYI, audio electronics is only my leisure hobby to supplement my classical music enjoyment. It’s kinda ‘peanut’ vs my EE experience.
My professional EE experience is much more subtantial:
providing engineering solutions to the power industries for deacades, e.g. to power companies for their HV power distribution.
I dealt with power company engineers day in day out.
Jack L
Deleted
I wouldn’t say that Stereophile is dying. In fact, of all the sites that I visit, Stereophile is first on my list. Sure, I visit Analogue planet, and stop by Twittering and maybe even Darko. But, my tastes were formed by the late Art Dudley. This is the first place I stop by to look for reviews of equipment that I am considering to purchase. Perhaps, I’m more in the maker end of the spectrum.
So, maybe a return to the idea of being a maker audiophile is in order. I’m certain my grandfather purchased your fist issues, because he had a subscription when he was 90 decades ago. He also designed audio equipment for RCA and such. So, I’m in the market for a SUT for an EMT TSD. I also just found out that EMT is making new tonearms! Yay!!! My next set of speakers I’ll be building with Rulliet drivers, but hey, I do well so a set of KEF LS60s might show up.
All in all, death looks somewhat premature.
I remain a steadfast fan.
We are definitely in the time of diminishing ‘hard copy’ circulation, but that is certainly not unique to Stereophile.
Plus, evil empires now rule the landscape, so I am even more impressed that Stereophile can maintain some degree of the appearance of autonomy.
Plus, the hobby is so different now.
In 1962, a person earning the median wage could participate in the hobby at the highest level. Now, it seems the average demo at a given hi fi show is double the median price for an entire home. That part might be a factor in how we are seen by non-audiophiles.
*Don’t get me wrong, I understand the great affordability of entry and middle ground gear. It’s just what JA1 called that price event horizon.
Thanks.
…the hobby is so different now. In 1962, a person earning the median wage could participate in the hobby at the highest level. Now, it seems the average demo at a given hi fi show is double the median price for an entire home. That part might be a factor in how we are seen by non-audiophiles.
*Don’t get me wrong, I understand the great affordability of entry and middle ground gear. It’s just what JA1 called that price event horizon.
*Don’t get me wrong, I understand the great affordability of entry and middle ground gear. It’s just what JA1 called that price event horizon.
I used that term in my January 2017 As We See It: www.stereophile.com/content/price-event-horizon. I was referring to an an earlier essay – www.stereophile.com/content/upward-price-spiral – in which I wrote “if an audio manufacturer has to gross a certain amount of revenue each quarter to meet payroll, cover fixed expenses, invest in parts for the next quarter’s production, and pay the interest on any capital they’ve borrowed, the least risky business strategy is to bring to market a very small number of very expensive products.” But that strategy make so many great products financially out of reach for regular people, like you and me, Anton.
John Atkinson
Technical Editor, Stereophile
I’m not complaining, but the Entry Level column needs to make a comeback.
Hi
Just last nite I streamed “Concert for George (Harrison)” performed 11/29/2002 at the Royal Albert Hall on my 4KUHD TV with sound feeding my audio rig.
What a full-house musical event! Paul McCartney & Ringo Starr were seen playing together with George’s then young son: Dhani, who looked so much like his dad !
Jack L
Hi
Though JGH outlived many audio celebrities back then (He was 79 !!), he could have lived much longer should he give up his addicted smoking. He only stopped smoking 10 years after he got the painful throat cancer !
This reflected his determination of running a readers-friendly audio magazine aginst those well established “showcases for advertisers”.
Jack L
A Salutation to the original JGH! A 2nd salutation to J.A. for his hard work and passion at the helm of Stereophile. May J.A.(2) have the same flair and zest for our beloved hobby. I have been a dedicated subscriber sine 1993 to present date.
Think about adding a “Dealers Showcase” several times per year to recognize those brick-and-mortar retailers still in the field.
Happy Listening!
J.A.(3)
60 years is an utterly amazing accomplishment! John’s work, Jim’s work, Larry’s… and of course Gordon, there has been no other magazine that provides the insight, knowledge and truth that Sterophile has provided over the past 6 decades!
I would even argue that they are the best, most interesting and best written magazine of ALL TIME.
To those that would say much less or worse about these fine writers and editors… you have no place to disparage with these professionals!
Thank you truly, for the most thought provoking periodical extant.
Wow! Remembering the 25th anniversary makes me feel old!
Congratulations to all. My 35 years with the magazine has provided tons of enjoyment, especially early on and for years when I read it ravenously in less than a day, then began the wait for the next issue (pre-internet!).
Bill
…to the team!! Pls keep up the great work!!
best regards
HB
I just upgraded from digital, TO PRINT, for the first time in the history of my subscription. A decadent move that will reward me in the most corporeal manner.
Long live Stereophile.
ALIBI Music Scares Up Trio of Horror Albums – Mix
Update your browser to view this website correctly. Update my browser now
×
Home The Wire
‘Minimal Horror,’ ‘Spooky Retro Horror’ and ‘Vocal Intrigue’ double-down on creepy in three very different ways.
By kristienb
Whether raising goosebumps, tingling spines or building uncomfortable suspense, music plays a powerful role in ratcheting up onscreen chills. That’s why ALIBI has dropped a trio of horror albums designed for maximum scares across a wide range of uses. The new production music releases are “Minimal Horror,” “Spooky Retro Horror” and “Vocal Intrigue” – each with its own inspiration and brand of creepy.
“Minimal Horror” (11 tracks, 456 audio files, Listen)
Inspired by trailers for “Midsommar,” “The Lodge,” “The Witch” and “Hereditary,” these stripped-back, building horror arrangements focus on mangled violins, scraped cellos, fluttering brass and twisted percussion. Huge wooshes and hits are interspersed with subtle, menacing ticking, tapping and scraping, while deep cellos and contra bass drone ominously. Many tracks also include designed vocals, such as guttural roars, tongue clicks and gurgling, and each has a gradual build to a nightmarish crescendo. A perfect match for horror trailers, survival horror video games and true crime documentaries.
“Spooky Retro Horror” (10 tracks, 440 audio files, Listen)
These sinister, building horror cues are inspired by iconic late-20th Century horror such as “Halloween,” “The Exorcist” and “The Omen,” to name a few. These tracks make use of a range of instrumentation, from creepy risers and stabbing staccato strings, to twinkling mallets, demonic chanting and ethereal, theremin-like female vocals for a sense of slow, creeping dread. Ideal if you’re cutting a deliberately retro horror scene, a Halloween spot or need some ominous music for your survival horror video game.
“Vocal Intrigue” (13 tracks, 595 audio files, Listen)
In this album, curious, intriguing and mysterious cues center around heavily stylized, rhythmic female vocal samples. Taking inspiration from iconic soundtracks such as “Squid Game,” this album develops the concept further by adding layers of trailer drums, ticking clocks, ominous sound design and menacing braams. The addition of staccato violins and cellos to some tracks also creates a hint of neoclassical elegance, making these tracks extremely well-suited to political intrigue/espionage, tense psychological thrillers, disturbing folk horror, survival terror and haunting puzzle/narrative games.
ALIBI Music’s horror cues have been used in the trailers promoting such notable projects as the Epix series “From,” Netflix original “Hellbound,” and the classic “Chucky” on USA/SyFy, among numerous others.
About ALIBI Music
Founded in 2011, ALIBI Music is a leading provider of music and sound effects for license in advertising, trailers, promos, programming, video games and all other forms of multimedia content. The company’s expertly curated catalogs are produced and structured especially for storytelling, with high-quality, searchable tracks, alternate mixes and stems delivering clients robust versatility. For more information, please visit www.alibimusic.com or connect via Twitter, LinkedIn, Facebook or Instagram.
alibi ⋅ Halloween ⋅ Horror ⋅ Music ⋅ production ⋅ Sound
kristienb
© 2022 MIX is part of Future plc, an international media group and leading digital publisher. Visit our corporate site.
