All issues

Newton and quasi-Newton methods of absolute optimization based on Cholesky factorization with adaptive step and finite difference approximation of the first and the second derivatives. In order to raise effectiveness of the quasi-Newton methods a modified version of Cholesky decomposition of quasi-Newton matrix is suggested. It solves the problem of step scaling while descending, allows approximation by non-quadratic functions, and integration with confidential neighborhood method. An approach to raise Newton methods effectiveness with finite difference approximation of the first and second derivatives is offered. The results of numerical research of algorithm effectiveness are shown.

In this paper, a time-dependent natural convective heat transfer in a closed square cavity filled with non- Newtonian fluid was considered in the presence of an isothermal energy source located on the lower wall of the region under consideration. The vertical boundaries were kept at constant low temperature, while the horizontal walls were completely insulated. The behavior of a non-Newtonian fluid was described by the Ostwald de Ville power law. The process under study was described by transient partial differential equations using dimensionless non-primitive variables “stream function – vorticity – temperature”. This method allows excluding the pressure field from the number of unknown parameters, while the non-dimensionalization allows generalizing the obtained results to a variety of physical formulations. The considered mathematical model with the corresponding boundary conditions was solved on the basis of the finite difference method. The algebraic equation for the stream function was solved by the method of successive lower relaxation. Discrete analogs of the vorticity equation and energy equation were solved by the Thomas algorithm. The developed numerical algorithm was tested in detail on a class of model problems and good agreement with other authors was achieved. Also during the study, the mesh sensitivity analysis was performed that allows choosing the optimal mesh.

As a result of numerical simulation of unsteady natural convection of a non-Newtonian power-law fluid in a closed square cavity with a local isothermal energy source, the influence of governing parameters was analyzed including the impact of the Rayleigh number in the range 104–106, power-law index $n = 0.6–1.4$, and also the position of the heating element on the flow structure and heat transfer performance inside the cavity. The analysis was carried out on the basis of the obtained distributions of streamlines and isotherms in the cavity, as well as on the basis of the dependences of the average Nusselt number. As a result, it was established that pseudoplastic fluids $(n < 1)$ intensify heat removal from the heater surface. The increase in the Rayleigh number and the central location of the heating element also correspond to the effective cooling of the heat source.

This article studies a method of constructing a predictive neural network model of a time series based on determining the composition of input variables, constructing a training sample and training itself using the back propagation method. Traditional methods of constructing predictive models of the time series are: the autoregressive model, the moving average model or the autoregressive model — the moving average allows us to approximate the time series by a linear dependence of the current value of the output variable on a number of its previous values. Such a limitation as linearity of dependence leads to significant errors in forecasting.

Mining Technologies using neural network modeling make it possible to approximate the time series by a nonlinear dependence. Moreover, the process of constructing of a neural network model (determining the composition of input variables, the number of layers and the number of neurons in the layers, choosing the activation functions of neurons, determining the optimal values of the neuron link weights) allows us to obtain a predictive model in the form of an analytical nonlinear dependence.

The determination of the composition of input variables of neural network models is one of the key points in the construction of neural network models in various application areas that affect its adequacy. The composition of the input variables is traditionally selected from some physical considerations or by the selection method. In this work it is proposed to use the behavior of the autocorrelation and private autocorrelation functions for the task of determining the composition of the input variables of the predictive neural network model of the time series.

In this work is proposed a method for determining the composition of input variables of neural network models for stationary and non-stationary time series, based on the construction and analysis of autocorrelation functions. Based on the proposed method in the Python programming environment are developed an algorithm and a program, determining the composition of the input variables of the predictive neural network model — the perceptron, as well as building the model itself. The proposed method was experimentally tested using the example of constructing a predictive neural network model of a time series that reflects energy consumption in different regions of the United States, openly published by PJM Interconnection LLC (PJM) — a regional network organization in the United States. This time series is non-stationary and is characterized by the presence of both a trend and seasonality. Prediction of the next values of the time series based on previous values and the constructed neural network model showed high approximation accuracy, which proves the effectiveness of the proposed method.

In this paper, we introduce a new version of Gauss–Newton method for solving a system of nonlinear equations based on ideas of the residual upper bound for a system of nonlinear equations and a quadratic regularization term. The introduced Gauss–Newton method in practice virtually forms the whole parameterized family of the methods solving systems of nonlinear equations and regression problems. The developed family of Gauss–Newton methods completely consists of iterative methods with generalization for cases of non-euclidean normed spaces, including special forms of Levenberg–Marquardt algorithms. The developed methods use the local model based on a parameterized proximal mapping allowing us to use an inexact oracle of «black–box» form with restrictions for the computational precision and computational complexity. We perform an efficiency analysis including global and local convergence for the developed family of methods with an arbitrary oracle in terms of iteration complexity, precision and complexity of both local model and oracle, problem dimensionality. We present global sublinear convergence rates for methods of the proposed family for solving a system of nonlinear equations, consisting of Lipschitz smooth functions. We prove local superlinear convergence under extra natural non-degeneracy assumptions for system of nonlinear functions. We prove both local and global linear convergence for a system of nonlinear equations under Polyak–Lojasiewicz condition for proposed Gauss– Newton methods. Besides theoretical justifications of methods we also consider practical implementation issues. In particular, for conducted experiments we present effective computational schemes for the exact oracle regarding to the dimensionality of a problem. The proposed family of methods unites several existing and frequent in practice Gauss–Newton method modifications, allowing us to construct a flexible and convenient method implementable using standard convex optimization and computational linear algebra techniques.

Modern methods of complex planning the execution of task packages in multistage systems are characterized by the presence of restrictions on the dimension of the problem being solved, the impossibility of guaranteed obtaining effective solutions for various values of its input parameters, as well as the impossibility of registration the conditions for the formation of sets from the result and the restriction on the interval duration of time of the system operating. The decomposition of the generalized function of the system into a set of hierarchically interconnected subfunctions is implemented to solve the problem of scheduling the execution of task packages with generating sets of results and the restriction on the interval duration of time for the functioning of the system. The use of decomposition made it possible to employ the hierarchical approach for planning the execution of task packages in multistage systems, which provides the determination of decisions by the composition of task groups at the first level of the hierarchy decisions by the composition of task packages groups executed during time intervals of limited duration at the second level and schedules for executing packages at the third level the hierarchy. In order to evaluate decisions on the composition of packages, the results of their execution, obtained during the specified time intervals, are distributed among the packages. The apparatus of the theory of hierarchical games is used to determine complex solutions. A model of a hierarchical game for making decisions by the compositions of packages, groups of packages and schedules of executing packages is built, which is a system of hierarchically interconnected criteria for optimizing decisions. The model registers the condition for the formation of sets from the results of the execution of task packages and restriction on duration of time intervals of its operating. The problem of determining the compositions of task packages and groups of task packages is NP-hard; therefore, its solution requires the use of approximate optimization methods. In order to optimize groups of task packages, the construction of a method for formulating initial solutions by their compositions has been implemented, which are further optimized. Moreover, a algorithm for distributing the results of executing task packages obtained during time intervals of limited duration by sets is formulated. The method of local solutions optimization by composition of packages groups, in accordance with which packages are excluded from groups, the results of which are not included in sets, and packages, that aren’t included in any group, is proposed. The software implementation of the considered method of complex optimization of the compositions of task packages, groups of task packages, and schedules for executing task packages from groups (including the implementation of the method for optimizing the compositions of groups of task packages) has been performed. With its use, studies of the features of the considered planning task are carried out. Conclusion are formulated concerning the dependence of the efficiency of scheduling the execution of task packages in multistage system under the introduced conditions from the input parameters of the problem. The use of the method of local optimization of the compositions of groups of task packages allows to increase the number of formed sets from the results of task execution in packages from groups by 60% in comparison with fixed groups (which do not imply optimization).

The paper is devoted to convex-concave saddle point problems where the objective is a sum of a large number of functions. Such problems attract considerable attention of the mathematical community due to the variety of applications in machine learning, including adversarial learning, adversarial attacks and robust reinforcement learning, to name a few. The individual functions in the sum usually represent losses related to examples from a data set. Additionally, the formulation admits a possibly nonsmooth composite term. Such terms often reflect regularization in machine learning problems. We assume that the dimension of one of the variable groups is relatively small (about a hundred or less), and the other one is large. This case arises, for example, when one considers the dual formulation for a minimization problem with a moderate number of constraints. The proposed approach is based on using Vaidya’s cutting plane method to minimize with respect to the outer block of variables. This optimization algorithm is especially effective when the dimension of the problem is not very large. An inexact oracle for Vaidya’s method is calculated via an approximate solution of the inner maximization problem, which is solved by the accelerated variance reduced algorithm Katyusha. Thus, we leverage the structure of the problem to achieve fast convergence. Separate complexity bounds for gradients of different components with respect to different variables are obtained in the study. The proposed approach is imposing very mild assumptions about the objective. In particular, neither strong convexity nor smoothness is required with respect to the low-dimensional variable group. The number of steps of the proposed algorithm as well as the arithmetic complexity of each step explicitly depend on the dimensionality of the outer variable, hence the assumption that it is relatively small.

This paper presents the results of experimental validation of some structural issues concerning the practical use of methods to overcome catastrophic forgetting of neural networks. A comparison of current effective methods like EWC (Elastic Weight Consolidation) and WVA (Weight Velocity Attenuation) is made and their advantages and disadvantages are considered. It is shown that EWC is better for tasks where full retention of learned skills is required on all the tasks in the training queue, while WVA is more suitable for sequential tasks with very limited computational resources, or when reuse of representations and acceleration of learning from task to task is required rather than exact retention of the skills. The attenuation of the WVA method must be applied to the optimization step, i. e. to the increments of neural network weights, rather than to the loss function gradient itself, and this is true for any gradient optimization method except the simplest stochastic gradient descent (SGD). The choice of the optimal weights attenuation function between the hyperbolic function and the exponent is considered. It is shown that hyperbolic attenuation is preferable because, despite comparable quality at optimal values of the hyperparameter of the WVA method, it is more robust to hyperparameter deviations from the optimal value (this hyperparameter in the WVA method provides a balance between preservation of old skills and learning a new skill). Empirical observations are presented that support the hypothesis that the optimal value of this hyperparameter does not depend on the number of tasks in the sequential learning queue. And, consequently, this hyperparameter can be picked up on a small number of tasks and used on longer sequences.

A model of combustion of premixed mixture of gases with one global chemical reaction is considered, the model includes equations of the second order for temperature of mixture and concentrations of fuel and oxidizer, and the right-hand sides of these equations contain the reaction rate function. This function depends on five unknown parameters of the global reaction and serves as approximation to multistep reaction mechanism. The model is reduced, after replacement of variables, to one equation of the second order for temperature of mixture that transforms to a first-order equation for temperature derivative depending on temperature that contains a parameter of flame propagation velocity. Thus, for computing the parameter of burning velocity, one has to solve Dirichlet problem for first-order equation, and after that a model dependence of burning velocity on mixture equivalence ratio at specified reaction rate parameters will be obtained. Given the experimental data of dependence of burning velocity on mixture equivalence ratio, the problem of optimal selection of reaction rate parameters is stated, based on minimization of the mean square deviation of model values of burning velocity on experimental ones. The aim of our study is analysis of uniqueness of this problem solution. To this end, we apply computational experiment during which the problem of global search of optima is solved using multistart of gradient descent. The computational experiment clarifies that the inverse problem in this statement is underdetermined, and every time, when running gradient descent from a selected starting point, it converges to a new limit point. The structure of the set of limit points in the five-dimensional space is analyzed, and it is shown that this set can be described with three linear equations. Therefore, it might be incorrect to tabulate all five parameters of reaction rate based on just one match criterion between model and experimental data of flame propagation velocity. The conclusion of our study is that in order to tabulate reaction rate parameters correctly, it is necessary to specify the values of two of them, based on additional optimality criteria.

Numerical study of unsteady mixed convection in an open partially porous horizontal channel with a heatgenerating source was performed. The outer surfaces of horizontal walls of finite thickness were adiabatic. In the channel there was a Newtonian heat-conducting fluid with a temperature-dependent viscosity. The discrete heatconducting and heat-generating source is located inside the bottom wall. The temperature of the fluid phase was equal to the temperature of the porous medium, and calculations were performed using the local thermal equilibrium model. The porous insertion is isotropic, homogeneous and permeable to fluid. The Darcy–Brinkman model was used to simulate the transport process within the porous medium. Governing equations formulated in dimensionless variables “stream function – vorticity – temperature” using the Boussinesq approximation were solved numerically by the finite difference method. The vorticity dispersion equation and energy equation were solved using locally one-dimensional Samarskii scheme. The diffusive terms were approximated by central differences, while the convective terms were approximated using monotonic Samarskii scheme. The difference equations were solved by the Thomas algorithm. The approximated Poisson equation for the stream function was solved separately by successive over-relaxation method. Optimal value of the relaxation parameter was found on the basis of computational experiments. The developed computational code was tested using a set of uniform grids and verified by comparing the results obtained of other authors.

Numerical analysis of unsteady mixed convection of variable viscosity fluid in the horizontal channel with a heat-generating source was performed for the following parameters: $\mathrm{Pr} = 7.0$, $\varepsilon = 0.8$, $\mathrm{Gr} = 10^5$, $C = 0-1$, $10^{-5} < \mathrm{Da} < 10^{-1}$, $50 < \mathrm{Re} < 500$, $\delta = l/H = 0.6-3$. Distributions of the isolines of the stream function, temperature and the dependences of the average Nusselt number and the average temperature inside the heater were obtained in a steady-state regime, when the stationary picture of the flow and heat transfer is observed. As a result we showed that an addition of a porous insertion leads to an intensification of heat removal from the surface of the energy source. The increase in the porous insertion sizes and the use of working fluid with different thermal characteristics, lead to a decrease in temperature inside the source.

Inverse geophysical problems are difficult to solve due to their mathematically incorrect formulation and large computational complexity. Geophysical exploration in frontier areas is even more complicated due to the lack of reliable geological information. In this case, inversion methods that allow interpretation of several types of geophysical data together are recognized to be of major importance. This paper is dedicated to one of such inversion methods, which is based on minimization of the determinant of the Gram matrix for a set of model vectors. Within the framework of this approach, we minimize a nonlinear functional, which consists of squared norms of data residual of different types, the sum of stabilizing functionals and a term that measures the structural similarity between different model vectors. We apply this approach to seismic and electromagnetic synthetic data set. Specifically, we study joint inversion of acoustic pressure response together with controlled-source electrical field imposing structural constraints on resulting electrical conductivity and P-wave velocity distributions.

We start off this note with the problem formulation and present the numerical method for inverse problem. We implemented the conjugate-gradient algorithm for non-linear optimization. The efficiency of our approach is demonstrated in numerical experiments, in which the true 3D electrical conductivity model was assumed to be known, but the velocity model was constructed during inversion of seismic data. The true velocity model was based on a simplified geology structure of a marine prospect. Synthetic seismic data was used as an input for our minimization algorithm. The resulting velocity model not only fit to the data but also has structural similarity with the given conductivity model. Our tests have shown that optimally chosen weight of the Gramian term may improve resolution of the final models considerably.