This study addresses inverse problems in exploration geophysics aimed at reconstructing the spatial distribution of subsurface medium properties from surface measurements of gravitational, magnetic, and electromagnetic fields. The integration of these geological techniques enhances the quality of solutions, but their practical implementation is limited by the availability of site-specific training data. To address this issue, we explore transfer learning approaches, comparing various neural network architectures and training configurations. Our findings indicate that optimized combinations of architecture and training outperform traditional methods, reducing reconstruction errors, maintaining reliability in noisy environments, and achieving comparable accuracy with smaller target training datasets.

This study addresses the problem of simultaneous determination of metal cations and nitrate anions concentrations in multicomponent aqueous solutions. To solve this task, a photoluminescent nanosensor based on carbon dots synthesized by the hydrothermal method from citric acid and ethylenediamine is developed. A key novel feature of the developed nanosensor is its multimodality, i.e. the ability to simultaneously determine (by the same photoluminescent spectrum) the concentrations of all ions under consideration. Various representations of the photoluminescent spectrum of carbon dots added to the studied solution are used as the source of information. To determine the target concentrations from the photoluminescent spectra under consideration, machine learning methods are used: neural networks of the multilayer perceptron type, convolutional neural networks, Kolmogorov-Arnold neural networks, gradient boosting, and linear regression. The use of the transfer learning technique for neural networks in the transition from solving a 6-parameter problem to solving a 7-parameter problem is considered. It was shown that the best results are achieved using convolutional neural networks. (The mean absolute errors of simultaneous determination of [Formula: see text], [Formula: see text], [Formula: see text], [Formula: see text], [Formula: see text], [Formula: see text] , cations and [Formula: see text] anion concentrations were 0.68, 1.05, 0.43, 1.38, 1.08, 0.32 and 2.43 mM, respectively.) Such precision meets the requirements for determining the concentration of ions in wastewater and process water. Transfer learning allows reducing the computational cost of the solution. Kolmogorov-Arnold networks can provide a visual interpretation of the resulting model.

This study addresses the solution of inverse problems in exploration geophysics using machine learning methods that involve reusing a trained model for different cases. At the same time, inverse problems are characterized by ill-posedness, for the reduction of which an approach is considered based on the indirect use of a priori information by taking it into account when forming a training sample through using narrow models of the media that describe some certain class of geological sections. However, this approach requires obtaining a separate solution for each case, with generation of a separate training dataset and training of a separate machine learning model (or set of models), which reduces usability and increases computational costs. Therefore, in order to reduce the computational cost, this study proposes to use transfer learning methods, which involve training base models on large datasets and fine-tuning them on limited datasets. This work is devoted to the study of the applicability of the transfer learning method in relation to inverse problems of gravimetry, magnetometry and magnetotelluric sounding, as well as their integration.

This study addresses the challenge of gas and volatile organic compound detection using semiconductor gas sensors, with a focus on mitigating the effects of sensor drift caused by aging and chemical degradation. Machine learning methods demonstrate high accuracy in gas concentration prediction within a single measurement series, achieving reliable regression performance. However, when applied to new data from different measurement series, the models suffer a complete loss of predictive ability due to inter-series variations. Traditional preprocessing methods, including PCA, fail to improve model transferability, highlighting the limitations of linear approaches. In contrast, nonlinear dimensionality reduction techniques—particularly autoencoder-based methods—show promise in identifying stable response patterns, leading to modest improvements in cross-series regression accuracy (especially for gases with long dynamic responses). Despite these advances, performance on independent series remains significantly inferior to within-series results, underscoring the need for further development of feature extraction methods. Future work should prioritize identifying invariant features in nonlinear latent spaces (e.g., autoencoder outputs) to enhance model robustness against sensor drift and inter-series variability.

This study addresses the problem of environmental monitoring of air in cities and industrial areas, which consists in detecting gases and volatile organic compounds using semiconductor gas sensors. To provide selectivity in the detection of certain gases, several semiconductor sensors with different doping components were tested. In addition, to ensure selectivity of gas determination, as well as high temporal resolution of the sensors, six types of nonlinear operating temperature conditions were used - the so-called heating dynamics. Due to the high complexity of the model describing the processes of interaction between gases and sensors, machine learning methods (linear regression with no regularization, lasso, ridge, random forest, gradient boosting and multilayer perceptron) based on the use of physical experiment data were used to process the sensor response. Optimal heating dynamics and optimal machine learning methods have been determined.

Kolmogorov-Arnold Networks (KAN), introduced in May 2024, are a novel type of artificial neural networks, whose abilities and properties are now being actively investigated by the machine learning community. In this study, we test application of KAN to solve an inverse problem for development of multimodal carbon luminescent nanosensors of ions dissolved in water, including heavy metal cations. We compare the results of solving this problem with four various machine learning methods—random forest, gradient boosting over decision trees, multi-layer perceptron neural networks, and KAN. Advantages and disadvantages of KAN are discussed, and it is demonstrated that KAN has high chance to become one of the algorithms most recommended for use in solving highly non-linear regression problems with moderate number of input features.

This study explores the feasibility of developing a dynamic cognovisor capable of recognizing cognitive states and transitions using fMRI data. Data were collected from 31 participants performing spatial and verbal tasks during fMRI scanning and were preprocessed using a nine-step algorithm for artifact removal and denoising. Three types of classification problems were examined, with machine learning methods and dimensionality reduction techniques applied to classify activity states. The best-performing models were identified for each classification problem, providing insights into their applicability. Notably, binary classification of resting versus active states achieved good quality with relatively simple methods. A key finding underscores the importance of accounting for temporal history of the signal prior to the prediction moment to improve model performance.

The study considers a virtual agent-based game environment “Stones”, designed to model and analyze interactions between agents of various types, including real people. The basic concept is very simple. The game features a game board with stones and agents that can move among them. A stone requires a specific number of agents to be removed; too few agents lack strength, while too many create confusion. The game continues until all stones are removed. On the other hand, it presents a wide scope for its complication through various modifications like movement modes, communication types, and win conditions. Therefore, it can be used as a benchmark to investigate the effectiveness of different agent architectures and training approaches, as well as in studies of the psychology of human behavior and interaction of humans with virtual agents. This study considers the simplest version of the gameplay with synchronous movement mode, no distance considerations, equivalence of all stones, no agent signals, and a cooperative game objective. Testing was conducted for algorithmic agents and emotional biologically inspired cognitive architecture (eBICA) agents. For eBICA agents, optimization of parameters was also performed using genetic algorithms.

The paper explores the possibilities of using data classification methods when forecasting time series of the geomagnetic Kp-index by machine learning methods. To classify categories of the Kp-index based on the degree of disturbance, linear and logistic regression, random forest, gradient boosting on top of decision trees, and artificial neural networks of various architectures are used. The results of these methods are compared with a trivial inertial forecast (the statistical indicators of which for problems of this type are always high) at horizons from 3 h to 1 day in 3-h increments. The problem of choosing a cross-validation scheme for selecting the model hyperparameters, ways to overcome the imbalance of categories, the relative importance of input features, as well as the dependence of the results on the test sample (beginning of the 25th solar activity cycle) on inclusion in the training sample of data from the 23rd and 24th cycles or only the 24th cycles are studied. Based on the results, conclusions are drawn about the preferred methods for classifying values of the Kp-index based on the level of geomagnetic disturbance. Ways for further research and possible improvement of the classification quality are outlined, including for determining the characteristic hidden states of Earth’s magnetosphere as a dynamic system in order to improve the quality of forecasting geomagnetic indices.

There are a lot of studies researching automated recognition of emotions. Emotions are represented as points in an emotion space. The emotion space itself is represented by different types of models. One is Facial Action Units System, another is Valence-Arousal-Dominance model. This study aims to create a mapping between these two emotion spaces. The data for the study was collected in a series of experiments with real humans, where both types of measurements were collected simultaneously. Given the data, we study the ability of machine learning models to create this type of mapping. We test different types of models against the task, such as tree-based models and linear models, and make conclusions about the optimal model.

Previously, it was shown that integration (joint use of data) of several geophysical methods allows one to obtain a higher quality of the solution of the inverse problem of exploration geophysics in comparison with the individual use of each of these methods. However, there may be a situation when for some measurement points there is no data from one of the geophysical methods used. At the same time, the data spaces of different integrated geophysical methods are interconnected. Therefore, the missing data of one method can be recovered from the known data of another one by constructing a preliminary adaptive mapping of one of the spaces to another. In this study, we investigate the solution of the inverse problem with integration of geophysical methods on the recovered data obtained based on noise addition during training of the neural networks performing the mapping from the data space of the method(s) with all data present to the data space of the method with missing data.

This article is devoted to the history of development and main research areas of the scientific school in the field of pattern recognition, image processing and analysis, and artificial intelligence and machine learning, founded in the early 1990s at the Skobeltsyn Institute of Nuclear Physics, Lomonosov Moscow State University (SINP MSU) by Prof. Igor’ Georgievich Persiantsev. For many years Persiantsev was the permanent leader of this scientific school; he laid down the basic principles and approaches to scientific research that still guide his disciples to this day. During this time, more than 30 people became students of Persiantsev’s school, who carried out scientific work under his leadership or under the leadership of his disciples, defended their candidate’s dissertations or diploma at the Faculty of Physics, Lomonosov Moscow State University. The article provides a brief historical background and an overview of the areas of research and major works published over more than 30 years (from 1992 to 2023) by Persiantsev and his disciples.

Among dimensional models of emotions, two- and three-dimensional are most popular, while the true dimension of affective space is a matter of debates. Here we study the inherent dimension of the emotion space represented in facial expressions, along with the mapping of electromyography (EMG) signals recorded from facial muscles to expressed emotions. For this purpose, an experiment was conducted with parallel EMG recording from three facial muscles (Zygomaticus Major, Corrugator Supercilii, and Masseter) and video registration of the face with automated emotion recognition from the video stream. Data analysis based on machine learning methods confirmed the 3D nature of the affective space (at least its part reflected in facial expressions). This result is consistent with the VAD and PAD models. Possibilities of accounting for complex, higher-order, or social emotions without introducing additional dimensions are discussed. The second finding of this study is the ability to reconstruct all three significant principal components of expressed affects using EMG signals recorded from three facial muscles with the help of machine learning.

This study is devoted to solving inverse problems of exploration geophysics, which consist in reconstructing the spatial distribution of the properties of the medium in the thickness of the earth from the geophysical fields measured on its surface. We consider the methods of gravimetry, magnetometry, and magnetotelluric sounding, as well as their integration, i.e. simultaneous use of data from several geophysical methods to solve the inverse problem. To implement such integration, in our previous studies we have proposed a parameterization scheme that describes a layered geophysical model with fixed layer properties, in which the determined parameters were the positions of the boundaries between the layers. In the present study, this parameterization scheme is complicated so that the properties of the layers vary from pattern to pattern in the data set. To improve the quality of neural network solution of the described inverse problem, we consider an approach based on the use of a priori information about the physical properties of the layers, in which this information is used directly as additional input features for the neural network.

In this study, to create a carbon dots-based multimodal nanosensor of metal ions, a new approach to solving the inverse problem of fluorescence spectroscopy is presented. The problem is to simultaneously determine the concentration of heavy metal ions Cr ^3+ , Ni ^2+ , Cu ^2+ , and nitrate anions NO ^-_3 in water by carbon dots (CDs) fluorescence spectra. A method of spectral data augmentation is proposed. It is based on the generation of excitation-emission matrices of CDs fluorescence from the noise vector using variational autoencoders and further determination of ion concentration corresponding to the generated matrices with convolutional neural networks. Implementing the proposed approach allowed reducing the mean absolute error in determining the concentration of ions by 60 % for Cr ^3+ , by 41 % for Ni ^2+ , by 62 % for Cu ^2+ , and by 48 % for NO ^-_3 .

This article presents the results of solving an inverse problem in spectroscopy using integration of optical spectroscopy methods. The studied inverse problem is determining the concentrations of heavy metal ions in multicomponent solutions by Raman spectra, infrared spectra and optical absorption spectra. It is shown that the joint use of data from various physical methods make it possible to reduce the error of spectroscopic determination of concentrations. If the integrated methods differ significantly by their accuracy, then their integration is not effective. These effects are observed using various machine learning methods: random forest, gradient boosting and artificial neural networks – multilayer perceptrons. A series of experiments with solutions based on river water are also performed to estimate the variability of the fluorescence of natural waters in Moscow. A significant increase in the error level relative to solutions prepared in distilled water is observed. This indicates the need to develop new methods to improve the quality of solution of the investigated problem for diagnostics of real river waters.

One of the methods for the analysis of complex spectral bands (especially for spectra of liquid objects) is their decomposition into a limited number of spectral curves with physically reasonable shapes (Gaussian, Lorentzian, Voigt, etc.). Subsequent analysis of the dependences of the parameters of these contours on some external conditions in which the spectra are obtained may reveal some regularities that bear information about the physical processes taking place in the object. The problem with the required decomposition is that such a decomposition in the presence of noise in spectra is an incorrect inverse problem. Therefore, this problem is often solved by advanced optimization methods that are less likely to become stuck in local minima, such as genetic algorithms (GA). In the conventional version of GA, all individuals are similar regarding the probabilities and implementation of the main genetic operators (crossover and mutation) and the procedure of selection. In their preceding studies, the authors tested the gender GA (GGA), where the individuals of the two genders differ in terms of the mutation probability (higher for males) and the selection procedures for crossover (with the number of crossovers limited for females). In this study, we introduce additional differences between the genders in the procedures of selection and mutation. The improved modification of GGA is tested by comparing the efficiency of the conventional GA, GGA, and three versions of GGA with and without subsequent gradient descent in solving the problems of decomposition of the Raman valence band of liquid water into Gaussian contours.

Exploration geophysics requires solving specific inverse problems — reconstructing the spatial distribution of the medium properties in the thickness of the earth from the geophysical fields measured on its surface. We consider inverse problems of gravimetry, magnetometry, magnetotelluric sounding, and their integration, which means simultaneous use of various geophysical fields to reconstruct the desired distribution. Integration requires the determined parameters for all the methods to be the same. This may be achieved by the spatial statement of the problem, in which the task is to determine the boundaries of geophysical objects. In our previous studies, we considered the parameterization scheme where the inverse problem was to determine the lower boundary of several geological layers. Each layer was characterized by variable values of the depth of the lower boundary along the section, and by fixed values of density, magnetization, and resistivity, both for the layer and over the entire dataset. It was demonstrated that the integration of geophysical methods provides significantly better results than the use of each of the methods separately. The present study considers an extended and more realistic model of data—a parameterization scheme with variable properties of the medium, both along each layer and over the dataset.

Magnetic storms can cause disruptions in the operation of radio communications, pipelines, power lines, and electrical networks, and they may possibly cause human health problems. Therefore, prediction of geomagnetic disturbances is of great practical value. Geomagnetic disturbances are usually described with the help of geomagnetic indices, including the planetary index K_p which is provided at a 3-h interval. The approach used in this study implies classifying geomagnetic disturbances according to the level of the K_p index. To do so, the whole range of the index values is divided into several intervals according to the degree of disturbance. The input data are time series of parameters of solar wind and interplanetary magnetic field, measured onboard spacecraft at the L1 Lagrange point between the Sun and the Earth, aa well as the value of the K_p index itself. To account for the ‘‘memory’’ of the time series, delay embedding of all the parameters is used—for each of the parameters, its several preceding values are taken into account. Additional preprocessing of the parameters is performed by calculating moving averages and other statistical indicators of the time series. To perform classification, various machine learning methods such as gradient boosting and artificial neural networks are used. The optimal values of the parameters of each method are determined by cross-validation, and pattern misbalance among the classes is partially reduced using the SMOTE technique. It is demonstrated that the suggested approach outperforms the trivial inertial model for all the values of the prediction horizon from 3 to 24 h (with a 3-h step). The most efficient preprocessing methods are described, as well as the best machine learning models.

This study compares several modifications of a gender genetic algorithm (GGA). Aside the difference between the genders in the probability of mutation, we introduce two additional modifications: different implementations of selection and different laws of dependence of the probability of mutation on gene number within a chromosome. We use four test optimization problems in spaces of various dimensions to compare conventional GA, conventional GGA, and GGA with the additional modifications implemented separately or together. It is demonstrated that the proposed additional modifications outperform conventional GA and conventional GGA in the achieved value of the fitness function, especially in high-dimensional spaces. With increase in the problem dimension, they degrade more slowly. Also, the new modifications prevent premature convergence of the algorithm.