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

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.

References in corpus (2)

• Topological and Dynamical Complexity of Random Neural Networks
• Heterogeneous connections induce oscillations in large scale networks

Cited by in corpus (14)

• Power-law statistics and universal scaling in the absence of criticality
• Transition to chaos in random neuronal networks
• Transition to chaos in random networks with cell-type-specific connectivity
• Optimal sequence memory in driven random networks
• Interacting diffusions on random graphs with diverging degrees: hydrodynamics and large deviations
• Dynamical Theory for Adaptive Systems
• Irreversibility in Non-reciprocal Chaotic Systems
• A local Echo State Property through the largest Lyapunov exponent
• Regular graphs maximize the variability of random neural networks
• Context-dependent representation in recurrent neural networks
• Hierarchy of chaotic dynamics in random modular networks
• Stationary-State Statistics of a Binary Neural Network Model with Quenched Disorder
• Universality of the mean-field equations of networks of Hopfield-like neurons
• Symmetries constrain dynamics in a family of balanced neural networks