arXiv subjects

Arsenii Onuchin

Publications and source records attributed to Arsenii Onuchin.

7 recordsLinked to original sources

arXiv2025

Iterative Ricci-Foster Curvature Flow with GMM-Based Edge Pruning: A Novel Approach to Community Detection

Community detection in complex networks is a fundamental problem, open to new approaches in various scientific settings. We introduce a novel community detection method, based on Ricci flow on graphs. Our technique iteratively updates edge weights (their metric lengths) according to their (combinatorial) Foster version of Ricci curvature computed from effective resistance distance between the nodes. The latter computation is known to be done by pseudo-inverting the graph Laplacian matrix. At that, our approach is alternative to one based on Ollivier-Ricci geometric flow for community detection on graphs, significantly outperforming it in terms of computation time. In our proposed method, iterations of Foster-Ricci flow that highlight network regions of different curvature -- are followed by a Gaussian Mixture Model (GMM) separation heuristic. That allows to classify edges into ''strong'' (intra-community) and ''weak'' (inter-community) groups, followed by a systematic pruning of the former to isolate communities. We benchmark our algorithm on synthetic networks generated from the Stochastic Block Model (SBM), evaluating performance with the Adjusted Rand Index (ARI). Our results demonstrate that proposed framework robustly recovers the planted community structure of SBM-s, establishing Ricci-Foster Flow with GMM-clustering as a principled and computationally effective new tool for network analysis, tested against alternative Ricci-Ollivier flow coupled with spectral clustering.

cs.SI↗

arXiv2025

Effective Stimulus Propagation in Neural Circuits: Driver Node Selection

Precise control of signal propagation in modular neural networks represents a fundamental challenge in computational neuroscience. We establish a framework for identifying optimal control nodes that maximize stimulus transmission between weakly coupled neural populations. Using spiking stochastic block model networks, we systematically compare driver node selection strategies - including random sampling and topology-based centrality measures (degree, betweenness, closeness, eigenvector, harmonic, and percolation centrality) - to determine minimal control inputs for achieving inter-population synchronization. Targeted stimulation of just 10-20% of the most central neurons in the source population significantly enhances spiking propagation fidelity compared to random selection. This approach yields a 64-fold increase in signal transfer efficiency at critical inter-module connection densities. These findings establish a theoretical foundation for precision neuromodulation in biological neural systems and neurotechnology applications.

q-bio.NC↗

arXiv2024

Localization transition in non-Hermitian systems depending on reciprocity and hopping asymmetry

We studied the single-particle Anderson localization problem for non-Hermitian systems on directed graphs. Random regular graph and various undirected standard random graph models were modified by controlling reciprocity and hopping asymmetry parameters. We found the emergence of left, biorthogonal and right localized states depending on both parameters and graph structure properties such as node degree $d$. For directed random graphs, the occurrence of biorthogonal localization near exceptional points is described analytically and numerically. The clustering of localized states near the center of the spectrum and the corresponding mobility edge for left and right states are shown numerically. Structural features responsible for localization, such as topologically invariant nodes or drains and sources, were also described. Considering the diagonal disorder, we observed the disappearance of localization dependence on reciprocity around $W \sim 20$ for a random regular graph $d=4$. With a small diagonal disorder, the average biorthogonal fractal dimension drastically reduces. Around $W \sim 5$ localization scars occur within the spectrum, alternating as vertical bands of clustering of left and right localized states.

cond-mat.dis-nn↗

arXiv2024

Multiple Neuronal Specializations Elicited By Socially Driven Recognition Of Food Odors

This study investigates the dynamics of non-spatial specializations in hippocampal place cells during exposure to novel environments. Hippocampal place cells, known for their role in spatial mapping, exhibit multi-modal responses to sensory cues. The research focuses on understanding how these cells adapt their specialization in response to novel stimuli, specifically examining non-spatial determinants such as odors and social interactions. Using a social-driven food odor recognition model in mice, the study records CA1 hippocampal neuron activity through miniscope imaging. The experimental design involves demonstrations of novel odors to mice, followed by observation sessions with food options. The analysis employs deep neural network tools for behavior tracking and the custom-developed INTENS software package for identifying neural specializations. Results indicate multiple specializations, particularly those related to odor, with differences observed between training and testing sessions. The findings suggest a temporal aspect to the formation of these specializations in novel conditions, necessitating further investigation for precise tracking.

q-bio.NC↗

arXiv2024

Communities in C.elegans connectome through the prism of non-backtracking walks

The fundamental relationship between the mesoscopic structure of neuronal circuits and organismic functions they subserve is one of the major challenges in contemporary neuroscience. Formation of structurally connected modules of neurons enacts the conversion from single-cell firing to large-scale behaviour of an organism, highlighting the importance of their accurate profiling in the data. While connectomes are typically characterized by significant sparsity of neuronal connections, recent advances in network theory and machine learning have revealed fundamental limitations of traditionally used community detection approaches in cases where the network is sparse. Here we studied the optimal community structure in the structural connectome of C.elegans, for which we exploited a non-conventional approach that is based on non-backtracking random walks, virtually eliminating the sparsity issue. In full agreement with the previous asymptotic results, we demonstrated that non-backtracking walks resolve the ground truth annotation into clusters on stochastic block models (SBM) with the size and density of the connectome better than the spectral methods related to simple random walks. Based on the cluster detectability threshold, we determined that the optimal number of modules in a recently mapped connectome of C.elegans is 10, which precisely corresponds to the number of isolated eigenvalues in the spectrum of the non-backtracking flow matrix. Broadly, our work provides a robust network-based framework to reveal mesoscopic structures in sparse connectomic datasets, paving way to further investigation of connectome mechanisms for different functions.

q-bio.NC↗

arXiv2023

Topological data analysis suggests human brain networks reconfiguration in the transition from a resting state to cognitive load

The functional network of the brain continually adapts to changing environmental demands. The environmental changes closely connect with changes of active cognitive processes. In recent years, the network approach has emerged as a promising method for analyzing the neurophysiological mechanisms that underlie psychological functions. The present study examines topological characteristics of functional brain networks in resting state and in cognitive load, provided by the execution of the Sternberg Item Recognition Paradigm (SIRP) based on electroencephalographic data. We propose that the topological properties of the functional networks in the human brain are distinct between cognitive load and resting state with higher integration in the networks during cognitive load. It was shown that topological features of functional connectomes strongly depend on the type of cognitive process performed by the subject and change in accordance with task change. The analysis also demonstrated that functional connectivity during working memory tasks showed a faster emergence of homology groups generators, supporting the idea of a relationship between the initial stages of working memory execution and an increase in faster network integration, with connector hubs playing a crucial role.

q-bio.NC↗

arXiv2022

Perspectives and constraints on neural network models of neurobiological processes

Artificial and natural neural network models are a new toolkit which could be potentially have been used for clarifying of complex brain functions. To attend this goal, such models need to be neurobiologically realistic. However, although neural networks have advanced keenly in recent decades their strict similarity in aspects of brain anatomy and physiology is imperfect. In this work we discuss different types of neural models, including localist, attractor and deep network models, and also identify aspects under which their biological credibility can be improved. These conditions range from the choice of neuron models and of mechanisms of synaptic plasticity and learning to implementation of inhibition and control, along with network architectures (modularity, connectivity). We highlight recent advances in biologically inspired neural network models and their constraints.

q-bio.NC↗