All issues

The COVID-19 pandemic has caused increased interest in mathematical models of the epidemic process, since only statistical analysis of morbidity does not allow medium-term forecasting in a rapidly changing situation.

Among the specific features of COVID-19 that need to be taken into account in mathematical models are the heterogeneity of the pathogen, repeated changes in the dominant variant of SARS-CoV-2, and the relative short duration of post-infectious immunity.

In this regard, solutions to a system of differential equations for a SIR class model with a heterogeneous duration of post-infectious immunity were analytically studied, and numerical calculations were carried out for the dynamics of the system with an average duration of post-infectious immunity of the order of a year.

For a SIR class model with a heterogeneous duration of post-infectious immunity, it was proven that any solution can be continued indefinitely in time in a positive direction without leaving the domain of definition of the system.

For the contact number $R_0 \leqslant 1$, all solutions tend to a single trivial stationary solution with a zero share of infected people, and for $R_0 > 1$, in addition to the trivial solution, there is also a non-trivial stationary solution with non-zero shares of infected and susceptible people. The existence and uniqueness of a non-trivial stationary solution for $R_0 > 1$ was proven, and it was also proven that it is a global attractor.

Also, for several variants of heterogeneity, the eigenvalues of the rate of exponential convergence of small deviations from a nontrivial stationary solution were calculated.

It was found that for contact number values corresponding to COVID-19, the phase trajectory has the form of a twisting spiral with a period length of the order of a year.

This corresponds to the real dynamics of the incidence of COVID-19, in which, after several months of increasing incidence, a period of falling begins. At the same time, a second wave of incidence of a smaller amplitude, as predicted by the model, was not observed, since during 2020–2023, approximately every six months, a new variant of SARS-CoV-2 appeared, which was more infectious than the previous one, as a result of which the new variant replaced the previous one and became dominant.

Deep learning’s power stems from complex architectures; however, these can lead to overfitting, where models memorize training data and fail to generalize to unseen examples. This paper proposes a novel probabilistic approach to mitigate this issue. We introduce two key elements: Truncated Log-Uniform Prior and Truncated Log-Normal Variational Approximation, and Automatic Relevance Determination (ARD) with Bayesian Deep Neural Networks (BDNNs). Within the probabilistic framework, we employ a specially designed truncated log-uniform prior for noise. This prior acts as a regularizer, guiding the learning process towards simpler solutions and reducing overfitting. Additionally, a truncated log-normal variational approximation is used for efficient handling of the complex probability distributions inherent in deep learning models. ARD automatically identifies and removes irrelevant features or weights within a model. By integrating ARD with BDNNs, where weights have a probability distribution, we achieve a variational bound similar to the popular variational dropout technique. Dropout randomly drops neurons during training, encouraging the model not to rely heavily on any single feature. Our approach with ARD achieves similar benefits without the randomness of dropout, potentially leading to more stable training.

To evaluate our approach, we have tested the model on two datasets: the Canadian Institute For Advanced Research (CIFAR-10) for image classification and a dataset of Macroscopic Images of Wood, which is compiled from multiple macroscopic images of wood datasets. Our method is applied to established architectures like Visual Geometry Group (VGG) and Residual Network (ResNet). The results demonstrate significant improvements. The model reduced overfitting while maintaining, or even improving, the accuracy of the network’s predictions on classification tasks. This validates the effectiveness of our approach in enhancing the performance and generalization capabilities of deep learning models.

Considered the application of physics-informed neural networks using multi layer perceptrons to solve Cauchy initial value problems in which the right-hand sides of the equation are continuous monotonically increasing, decreasing or oscillating functions. With the use of the computational experiments the influence of the construction of the approximate neural network solution, neural network structure, optimization algorithm and software implementation means on the learning process and the accuracy of the obtained solution is studied. The analysis of the efficiency of the most frequently used machine learning frameworks in software development with the programming languages Python and C# is carried out. It is shown that the use of C# language allows to reduce the time of neural networks training by 20–40%. The choice of different activation functions affects the learning process and the accuracy of the approximate solution. The most effective functions in the considered problems are sigmoid and hyperbolic tangent. The minimum of the loss function is achieved at the certain number of neurons of the hidden layer of a single-layer neural network for a fixed training time of the neural network model. It’s also mentioned that the complication of the network structure increasing the number of neurons does not improve the training results. At the same time, the size of the grid step between the points of the training sample, providing a minimum of the loss function, is almost the same for the considered Cauchy problems. Training single-layer neural networks, the Adam method and its modifications are the most effective to solve the optimization problems. Additionally, the application of twoand three-layer neural networks is considered. It is shown that in these cases it is reasonable to use the LBFGS algorithm, which, in comparison with the Adam method, in some cases requires much shorter training time achieving the same solution accuracy. The specificity of neural network training for Cauchy problems in which the solution is an oscillating function with monotonically decreasing amplitude is also investigated. For these problems, it is necessary to construct a neural network solution with variable weight coefficient rather than with constant one, which improves the solution in the grid cells located near by the end point of the solution interval.

This paper addresses the problem of detecting small objects in visible-spectrum aerial imagery. High background variability and weak feature saliency make small object detection a non-trivial task. Under such conditions, classical computer vision algorithms based on hand-crafted descriptors exhibit low efficiency, prompting a shift towards deep neural network architectures, which demonstrate superior generalization capability and robustness to false positives. We selected a one-stage approach based on a neural network predictive model as the primary object detection method. Also, several neural network architectures belonging to this class were analyzed, outlining their advantages and disadvantages. As the baseline detector, we adopted YOLO version 11 and incorporated a multi-scale feature aggregation module (which combines information from neural network layers operating at different scales) and a dimension-aware selective integration module (which automatically determines the dimension (channel, height, or width) along which to process features and fuses them selectively). These modifications aim to both enhance computational efficiency, enabling deployment of neural network models onboard aerial vehicles for real-time image analysis, and improve small object detection accuracy. Given the complexity of image annotation, we used an open synthetic image database containing approximately 4 000 images for training and testing (with a 90/10 split, respectively). We extend the training set using various random augmentation techniques. To evaluate the performance of the resulting predictive models, we employed mean Average Precision across all classes, using both a fixed 50% intersectionover- union threshold and a varying threshold from 50% to 95%. Overfitting was monitored by analyzing loss curves during training process. The proposed modifications to the YOLO architecture reduced image processing time by a factor of two while maintaining detection accuracy.

Work is devoted to modeling instrumental errors of a laser beam diameter measurement using a method based on a lambertian transmissive screen. Super-Lorenz distribution was used as a model of the beam. To determine the effect of each parameter on the measurement error were performed computational experiments, results of which were approximated by analytic functions. There were obtained the errors depending on relative beam size, spatial non-uniformity of the transmission screen, lens distortion, physical vignetting, beam tilt, CCD spatial resolution, ADC resolution of a camera. There was shown that the error can be less then 1 %.

This paper formulates and solves a practical problem of data recovery regarding the distribution of languages on regional level in context of China. The necessity of this recovery is related to the problem of the determination of the linguistic diversity indices, which, in turn, are used to analyze empirically and to predict sources of social and economic development as well as to indicate potential conflicts at regional level. We use Ethnologue database and China census as the initial data sources. For every language spoken in China, the data contains (a) an estimate of China residents who claim this language to be their mother tongue, and (b) indicators of the presence of such residents in China provinces. For each pair language/province, we aim to estimate the number of the province inhabitants that claim the language to be their mother tongue. This base problem is reduced to solving an undetermined system of algebraic equations. Given additional restriction that Ethnologue database introduces data collected at different time moments because of gaps in Ethnologue language surveys and accompanying data collection expenses, we relate those data to a single time moment, that turns the initial task to an ’ill-posed’ system of algebraic equations with imprecisely determined right hand side. Therefore, we are looking for an approximate solution characterized by a minimal discrepancy of the system. Since some languages are much less distributed than the others, we minimize the weighted discrepancy, introducing weights that are inverse to the right hand side elements of the equations. This definition of discrepancy allows to recover the required variables. More than 92% of the recovered variables are robust to probabilistic modelling procedure for potential errors in initial data.

In case of vessel wall damage or contact of blood plasma with a foreign surface, the chain of chemical reactions called coagulation cascade is launched that leading to the formation of a fibrin clot. A key enzyme of the coagulation cascade is thrombin, which catalyzes formation of fibrin from fibrinogen. The distribution of thrombin concentration in blood plasma determines spatio-temporal dynamics of clot formation. Contact pathway of blood coagulation triggers the production of thrombin in response to the contact with a negatively charged surface. If the concentration of thrombin generated at this stage is large enough, further production of thrombin takes place due to positive feedback loops of the coagulation cascade. As a result, thrombin propagates in plasma cleaving fibrinogen that results in the clot formation. The concentration profile and the speed of propagation of thrombin are constant and do not depend on the type of the initial activator.

Such behavior of the coagulation system is well described by the traveling wave solutions in a system of “reaction – diffusion” equations on the concentration of blood factors involved in the coagulation cascade. In this study, we carried out detailed analysis of the mathematical model describing the main reaction of the intrinsic pathway of coagulation cascade.We formulate necessary and sufficient conditions of the existence of the traveling wave solutions. For the considered model the existence of such solutions is equivalent to the existence of the wave solutions in the simplified one-equation model describing the dynamics of thrombin concentration derived under the quasi-stationary approximation.

Simplified model also allows us to obtain analytical estimate of the thrombin propagation rate in the considered model. The speed of the traveling wave for one equation is estimated using the narrow reaction zone method and piecewise linear approximation. The resulting formulas give a good approximation of the velocity of propagation of thrombin in the simplified, as well as in the original model.

Financial mathematics is one of the most natural applications for the statistical analysis of time series. Financial time series reflect simultaneous activity of a large number of different economic agents. Consequently, one expects that methods of statistical physics and the theory of random processes can be applied to them.

In this paper, we provide a statistical analysis of time series of the FOREX currency market. Of particular interest is the comparison of the time series behavior depending on the way time is measured: physical time versus trading time measured in the number of elementary price changes (ticks). The experimentally observed statistics of the time series under consideration (euro–dollar for the first half of 2007 and for 2009 and British pound – dollar for 2007) radically differs depending on the choice of the method of time measurement. When measuring time in ticks, the distribution of price increments can be well described by the normal distribution already on a scale of the order of ten ticks. At the same time, when price increments are measured in real physical time, the distribution of increments continues to differ radically from the normal up to scales of the order of minutes and even hours.

To explain this phenomenon, we investigate the statistical properties of elementary increments in price and time. In particular, we show that the distribution of time between ticks for all three time series has a long (1-2 orders of magnitude) power-law tails with exponential cutoff at large times. We obtained approximate expressions for the distributions of waiting times for all three cases. Other statistical characteristics of the time series (the distribution of elementary price changes, pair correlation functions for price increments and for waiting times) demonstrate fairly simple behavior. Thus, it is the anomalously wide distribution of the waiting times that plays the most important role in the deviation of the distribution of increments from the normal. As a result, we discuss the possibility of applying a continuous time random walk (CTRW) model to describe the FOREX time series.

The biological structure is considered as an open nonequilibrium system which properties can be described on the basis of kinetic equations. New problems with nonequilibrium boundary conditions are introduced. The nonequilibrium distribution tends gradually to an equilibrium state. The region of spatial inhomogeneity has a scale depending on the rate of mass transfer in the open system and the characteristic time of metabolism. In the proposed approximation, the internal energy of the motion of molecules is much less than the energy of translational motion. Or in other terms we can state that the kinetic energy of the average blood velocity is substantially higher than the energy of chaotic motion of the same particles. We state that the relaxation problem models a living system. The flow of entropy to the system decreases in downstream, this corresponds to Shrödinger’s general ideas that the living system “feeds on” negentropy. We introduce a quantity that determines the complexity of the biosystem, more precisely, this is the difference between the nonequilibrium kinetic entropy and the equilibrium entropy at each spatial point integrated over the entire spatial region. Solutions to the problems of spatial relaxation allow us to estimate the size of biosystems as regions of nonequilibrium. The results are compared with empirical data, in particular, for mammals we conclude that the larger the size of animals, the smaller the specific energy of metabolism. This feature is reproduced in our model since the span of the nonequilibrium region is larger in the system where the reaction rate is shorter, or in terms of the kinetic approach, the longer the relaxation time of the interaction between the molecules. The approach is also used for estimation of a part of a living system, namely a green leaf. The problems of aging as degradation of an open nonequilibrium system are considered. The analogy is related to the structure, namely, for a closed system, the equilibrium of the structure is attained for the same molecules while in the open system, a transition occurs to the equilibrium of different particles, which change due to metabolism. Two essentially different time scales are distinguished, the ratio of which is approximately constant for various animal species. Under the assumption of the existence of these two time scales the kinetic equation splits in two equations, describing the metabolic (stationary) and “degradative” (nonstationary) parts of the process.

This paper studies electromagnetic fields, frequencies of lasing, and emission thresholds of a drop-shaped microcavity laser. From the mathematical point of view, the original problem is a nonstandard two-parametric eigenvalue problem for the Helmholtz equation on the whole plane. The desired positive parameters are the lasing frequency and the threshold gain, the corresponding eigenfunctions are the amplitudes of the lasing modes. This problem is usually referred to as the lasing eigenvalue problem. In this study, spectral characteristics are calculated numerically, by solving the lasing eigenvalue problem on the basis of the set of Muller boundary integral equations, which is approximated by the Nystr¨om method. The Muller equations have weakly singular kernels, hence the corresponding operator is Fredholm with zero index. The Nyström method is a special modification of the polynomial quadrature method for boundary integral equations with weakly singular kernels. This algorithm is accurate for functions that are well approximated by trigonometric polynomials, for example, for eigenmodes of resonators with smooth boundaries. This approach leads to a characteristic equation for mode frequencies and lasing thresholds. It is a nonlinear algebraic eigenvalue problem, which is solved numerically by the residual inverse iteration method. In this paper, this technique is extended to the numerical modeling of microcavity lasers having a more complicated form. In contrast to the microcavity lasers with smooth contours, which were previously investigated by the Nyström method, the drop has a corner. We propose a special modification of the Nyström method for contours with corners, which takes also the symmetry of the resonator into account. The results of numerical experiments presented in the paper demonstrate the practical effectiveness of the proposed algorithm.