Lev Shchur
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Lev Shchur has a total of 54 co-authors. Here are the top co-authors by number of shared publications:
- Evgeni Burovski
HSE University
7 shared publications - Bertrand Berche
Laboratoire de Physique et Chimie Théoriques, CNRS - Université de Lorraine, UMR 2 , 7019 Nancy
5 shared publications - Vladislav Chertenkov
Landau Institute for Theoretical Physics
5 shared publications - Maria Guskova
Landau Institute for Theoretical Physics
4 shared publications - Diana Sukhoverkhova
Landau Institute for Theoretical Physics
3 shared publications - Liliia Ziganurova
Landau Institute for Theoretical Physics, Chernogolovka, Russia
3 shared publications
Latest Publications
- Open Access
- 30 October 2026
- Published by Elsevier in Nuclear Physics B
- Vol. 1033
Algorithmic overlaps in the Baxter–Wu model: Cluster dynamics under Novotny–Evertz updates
- 30 October 2026
- Published by Elsevier in Nuclear Physics B
- Vol. 1033
Open AccessAlgorithmic overlaps in the Baxter–Wu model: Cluster dynamics under Novotny–Evertz updates
- Open Access
- 28 September 2026
- Published by Springer Nature
- p. 145-155
On the Validity of Transfer Learning for Temperature and Critical Exponent Extraction
We analyze the Ising model with anisotropic interactions in two dimensions using deep learning neural networks (NN) trained to classify paramagnetic (PM) and ferromagnetic (FM) phases. The spin configurations used as training data are labeled as belonging to either PM state or the FM state using analytically known phase transition temperatures. We use the variation of the output function to determine the critical temperature and to estimate the correlation length critical exponent, $$\nu $$ , in a wide range of the anisotropy parameter. We find that on increasing the anisotropy, the critical exponent remains consistent with the expected value $$\nu {=}1$$ for the Ising universality class. At the same time, the NN prediction for the critical temperature, $$T_c$$ , is shifted from the analytically known value, with the magnitude of the shift increasing with anisotropy.- 28 September 2026
- Published by Springer Nature
- p. 145-155
Open AccessOn the Validity of Transfer Learning for Temperature and Critical Exponent Extraction
We analyze the Ising model with anisotropic interactions in two dimensions using deep learning neural networks (NN) trained to classify paramagnetic (PM) and ferromagnetic (FM) phases. The spin configurations used as training data are labeled as belonging to either PM state or the FM state using analytically known phase transition temperatures. We use the variation of the output function to determine the critical temperature and to estimate the correlation length critical exponent, $$\nu $$ , in a wide range of the anisotropy parameter. We find that on increasing the anisotropy, the critical exponent remains consistent with the expected value $$\nu {=}1$$ for the Ising universality class. At the same time, the NN prediction for the critical temperature, $$T_c$$ , is shifted from the analytically known value, with the magnitude of the shift increasing with anisotropy. - 12 June 2026
- Published by American Physical Society (APS) in Physical Review B
- Vol. 114 (1)
Algorithmic overlaps as thermodynamic variables: From local to cluster Monte Carlo dynamics in critical phenomena
- 12 June 2026
- Published by American Physical Society (APS) in Physical Review B
- Vol. 114 (1)
Algorithmic overlaps as thermodynamic variables: From local to cluster Monte Carlo dynamics in critical phenomena
- Open Access
- 21 May 2026
- Published by Elsevier in Physica A: Statistical Mechanics and its Applications
- Vol. 696
Ising models on the hydrogen peroxide and other lattices
- 21 May 2026
- Published by Elsevier in Physica A: Statistical Mechanics and its Applications
- Vol. 696
Open AccessIsing models on the hydrogen peroxide and other lattices
- 16 April 2026
- Published by World Scientific Pub Co Pte Ltd in International Journal of Bifurcation and Chaos
- Vol. 36 (10)
Cross-Influence of Two Societies in Deterministic Evolutionary Game
We present a simple model that simulates the possible influence of one society on another. Specifically, two societies evolve deterministically according to the well-known Nowak–May spatial game with the addition of mutual influence through connections that reflect the current states of the societies. This may be related to the influence of a global information resource on the society. We calculate persistence as a characteristic of chaotic behavior. A particularly interesting and novel phenomenon is that the quasi-stationary state of one society can lead to chaotic behavior of its individual members under the influence of another society, while the society itself remains stable.
- 16 April 2026
- Published by World Scientific Pub Co Pte Ltd in International Journal of Bifurcation and Chaos
- Vol. 36 (10)
Cross-Influence of Two Societies in Deterministic Evolutionary Game
We present a simple model that simulates the possible influence of one society on another. Specifically, two societies evolve deterministically according to the well-known Nowak–May spatial game with the addition of mutual influence through connections that reflect the current states of the societies. This may be related to the influence of a global information resource on the society. We calculate persistence as a characteristic of chaotic behavior. A particularly interesting and novel phenomenon is that the quasi-stationary state of one society can lead to chaotic behavior of its individual members under the influence of another society, while the society itself remains stable.
- 17 October 2025
- Published by American Physical Society (APS) in Physical Review E
- Vol. 112 (4) , 044128
Phase probabilities in first-order transitions using machine learning
We set out to explore the possibility of investigating the critical behavior of systems with first-order phase transition using deep machine learning. We propose a machine learning protocol with ternary classification of instantaneous spin configurations using known values of disordered phase energy and ordered phase energy. The trained neural network is used to predict whether a given sample belong to one or another phase of matter. This allows us to estimate for the first time the probability that configurations with a certain energy belong to the ordered phase, coexistence phase, and disordered phase. Based on these probabilities, we obtained estimates of the values of the critical energies and latent heat for the Potts model with 10 and 20 components, which undergoes a strong discontinuous transition. We also found that the probabilities may reflect geometric transitions in the coexistence phase.
- 17 October 2025
- Published by American Physical Society (APS) in Physical Review E
- Vol. 112 (4) , 044128
Phase probabilities in first-order transitions using machine learning
We set out to explore the possibility of investigating the critical behavior of systems with first-order phase transition using deep machine learning. We propose a machine learning protocol with ternary classification of instantaneous spin configurations using known values of disordered phase energy and ordered phase energy. The trained neural network is used to predict whether a given sample belong to one or another phase of matter. This allows us to estimate for the first time the probability that configurations with a certain energy belong to the ordered phase, coexistence phase, and disordered phase. Based on these probabilities, we obtained estimates of the values of the critical energies and latent heat for the Potts model with 10 and 20 components, which undergoes a strong discontinuous transition. We also found that the probabilities may reflect geometric transitions in the coexistence phase.
- 2 September 2025
- Published by American Physical Society (APS) in Physical Review E
- Vol. 112 (3) , 034104
Machine learning domain adaptation in spin models with continuous phase transitions
The main question raised in the article is whether a neural network trained on a spin lattice model in one universality class can be used to test a model in another universality class. The quantities of interest are the critical phase transition temperature and the correlation length exponent. In other words, the question of transfer learning is how "universal" the trained network is and under what conditions. For this purpose, we applied a supervised learning procedure to three two-dimensional models for which critical properties are precisely known: The Ising model, the four-state Potts model, and the Baxter-Wu model. We consider two datasets: One with spin configurations and one with binding energy configurations. We find that estimates of the critical temperature agree well with the known results for both datasets, but not with the results of cross-testing using the energy datasets of the two models: The four-state Potts model and the Ising model. Estimates of the critical length exponent are less regular, and appear to be more accurate for energy datasets. A good example is the cross-testing using the energy dataset between the Ising model and the Baxter-Wu model in both training and testing directions.
- 2 September 2025
- Published by American Physical Society (APS) in Physical Review E
- Vol. 112 (3) , 034104
Machine learning domain adaptation in spin models with continuous phase transitions
The main question raised in the article is whether a neural network trained on a spin lattice model in one universality class can be used to test a model in another universality class. The quantities of interest are the critical phase transition temperature and the correlation length exponent. In other words, the question of transfer learning is how "universal" the trained network is and under what conditions. For this purpose, we applied a supervised learning procedure to three two-dimensional models for which critical properties are precisely known: The Ising model, the four-state Potts model, and the Baxter-Wu model. We consider two datasets: One with spin configurations and one with binding energy configurations. We find that estimates of the critical temperature agree well with the known results for both datasets, but not with the results of cross-testing using the energy datasets of the two models: The four-state Potts model and the Ising model. Estimates of the critical length exponent are less regular, and appear to be more accurate for energy datasets. A good example is the cross-testing using the energy dataset between the Ising model and the Baxter-Wu model in both training and testing directions.
- 1 March 2025
- Published by Pleiades Publishing Ltd in Lobachevskii Journal of Mathematics
- Vol. 46 (3) , 1315-1321
Generation of Artificial Images of Cross Sections of WC/Co Composite Alloys Using Diffusion Networks
The study of statistical properties of microstructures of composite materials is carried out by analysing microphotographs of material cuts. The obtained two-dimensional structure represents cuts of composite elements, the geometrical properties of which can be studied by computer vision methods. Hundreds of micro-images can be collected from one microstructure, which allows to study statistical properties of mutual arrangement of composite elements. To increase sampling, the possibility of creating artificial microstructures using diffusion neural networks is being investigated. The plausibility criterion is the comparison of corresponding distributions for real and artificial microstructures, in particular using the Wasserstein–Kantorovich–Rubinstein distance.
- 1 March 2025
- Published by Pleiades Publishing Ltd in Lobachevskii Journal of Mathematics
- Vol. 46 (3) , 1315-1321
Generation of Artificial Images of Cross Sections of WC/Co Composite Alloys Using Diffusion Networks
The study of statistical properties of microstructures of composite materials is carried out by analysing microphotographs of material cuts. The obtained two-dimensional structure represents cuts of composite elements, the geometrical properties of which can be studied by computer vision methods. Hundreds of micro-images can be collected from one microstructure, which allows to study statistical properties of mutual arrangement of composite elements. To increase sampling, the possibility of creating artificial microstructures using diffusion neural networks is being investigated. The plausibility criterion is the comparison of corresponding distributions for real and artificial microstructures, in particular using the Wasserstein–Kantorovich–Rubinstein distance.
- 31 January 2025
- Published by Springer Nature
- p. 434-449
Supervised and Transfer Learning for Phase Transition Research
- 31 January 2025
- Published by Springer Nature
- p. 434-449
Supervised and Transfer Learning for Phase Transition Research
- 1 January 2025
- Published by Pleiades Publishing Ltd in Lobachevskii Journal of Mathematics
- Vol. 46 (1) , 528-534
Transfer Machine Learning of an Anisotropic Model
We investigate the possibility of extracting features of second-order phase transitions using transfer machine learning. We have performed supervised machine learning for binary classification of snapshots of the spin distribution of the isotropic Ising model. The binary classification is performed in ferromagnetic and paramagnetic phases using a known critical temperature. The trained network is used to predict whether a snapshot obtained from model simulations with orthogonal anisotropy belongs to the paramagnetic phase. Using finite-dimensional prediction analysis, we estimate the critical temperature and the exponent of the correlation length. This gives us an estimate of the interval of the anisotropy parameter in which the neural network can make correct estimates.
- 1 January 2025
- Published by Pleiades Publishing Ltd in Lobachevskii Journal of Mathematics
- Vol. 46 (1) , 528-534
Transfer Machine Learning of an Anisotropic Model
We investigate the possibility of extracting features of second-order phase transitions using transfer machine learning. We have performed supervised machine learning for binary classification of snapshots of the spin distribution of the isotropic Ising model. The binary classification is performed in ferromagnetic and paramagnetic phases using a known critical temperature. The trained network is used to predict whether a snapshot obtained from model simulations with orthogonal anisotropy belongs to the paramagnetic phase. Using finite-dimensional prediction analysis, we estimate the critical temperature and the exponent of the correlation length. This gives us an estimate of the interval of the anisotropy parameter in which the neural network can make correct estimates.