Large deviations, dynamics and phase transitions in large stochastic and disordered neural networks

DOI10.1007/S10955-013-0818-5zbMATH Open1291.82099arXiv1302.6951OpenAlexW2257652304WikidataQ105583419 ScholiaQ105583419MaRDI QIDQ2016499FDOQ2016499

Jonathan D. Touboul, Tanguy Cabana

Publication date: 20 June 2014

Published in: Journal of Statistical Physics (Search for Journal in Brave)

Abstract: We analyze the macroscopic behavior of multi-populations randomly connected neural networks with interaction delays. Similar to cases occurring in spin glasses, we show that the sequences of empirical measures satisfy a large deviation principle, and converge towards a self-consistent non-Markovian process. The proof differs in that we are working in infinite-dimensional spaces (interaction delays), non-centered interactions and multiple cell types. The limit equation is qualitatively analyzed, and we identify a number of phase transitions in such systems upon changes in delays, connectivity patterns and dispersion, particularly focusing on the emergence of non-equilibrium states involving synchronized oscillations.

Full work available at URL: https://arxiv.org/abs/1302.6951

Recommendations

• Large deviations for randomly connected neural networks. I: Spatially extended systems
• Large deviations for randomly connected neural networks. II: State-dependent interactions
• Large deviations for nonlocal stochastic neural fields
• On the dynamics of random neuronal networks
• Statistical physics of interacting neural networks
• Dynamics of stochastic neuronal networks and the connections to random graph theory
• Large deviations and mean-field theory for asymmetric random recurrent neural networks
• Phase transitions in the generalization behaviour of multilayer neural networks

zbMATH Keywords

large deviationsphase transitionsmean-field equationsheterogeneous neuronal networks

Mathematics Subject Classification ID

Transition functions, generators and resolvents (60J35) Phase transitions (general) in equilibrium statistical mechanics (82B26) Disordered systems (random Ising models, random Schrödinger operators, etc.) in equilibrium statistical mechanics (82B44) Neural nets applied to problems in time-dependent statistical mechanics (82C32)

Cites Work

• Near-Optimal Signal Recovery From Random Projections: Universal Encoding Strategies?
• Introduction to functional differential equations
• Title not available (Why is that?)
• Stochastic Equations in Infinite Dimensions
• Neural networks and physical systems with emergent collective computational abilities.
• Large deviations techniques and applications.
• Spatiotemporal dynamics of continuum neural fields
• Characteristics of Random Nets of Analog Neuron-Like Elements
• Large deviations for some weakly dependent random processes
• Propagation of chaos in neural fields
• �quations de type de Boltzmann, spatialement homog�nes
• Title not available (Why is that?)
• Mean-field equations for stochastic firing-rate neural fields with delays: derivation and noise-induced transitions
• Title not available (Why is that?)
• Noise-Induced Behaviors in Neural Mean Field Dynamics
• Prescribing a System of Random Variables by Conditional Distributions
• Systematic Fluctuation Expansion for Neural Network Activity Equations
• Large deviations and mean-field theory for asymmetric random recurrent neural networks
• Large deviations for Langevin spin glass dynamics
• Mean-field equations, bifurcation map and route to chaos in discrete time neural networks
• Averaged and quenched propagation of chaos for spin glass dynamics
• Mean-field theory and synchronization in random recurrent neural networks

Cited In (13)

• Interacting diffusions on sparse graphs: hydrodynamics from local weak limits
• Interacting diffusions on random graphs with diverging average degrees: hydrodynamics and large deviations
• Universality for Langevin-like spin glass dynamics
• Symmetries constrain dynamics in a family of balanced neural networks
• An emergent autonomous flow for mean-field spin glasses
• Phase transitions in the neuropercolation model of neural populations with mixed local and non-local interactions
• Large Population Asymptotics for Interacting Diffusions in a Quenched Random Environment
• Mathematical studies of the dynamics of finite-size binary neural networks: a review of recent progress
• Large deviations for nonlocal stochastic neural fields
• Title not available (Why is that?)
• Large deviations for randomly connected neural networks: I. Spatially extended systems
• Large deviations for randomly connected neural networks: II. State-dependent interactions
• Limits and dynamics of stochastic neuronal networks with random heterogeneous delays

This page was built for publication: Large deviations, dynamics and phase transitions in large stochastic and disordered neural networks

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2016499)