An optimization-based equilibrium measure describes non-equilibrium steady state dynamics: application to edge of chaos

Abstract

Understanding neural dynamics is a central topic in machine learning, non-linear physics and neuroscience. However, the dynamics is non-linear, stochastic and particularly non-gradient, i.e., the driving force can not be written as gradient of a potential. These features make analytic studies very challenging. The common tool is the path integral approach or dynamical mean-field theory, but the drawback is that one has to solve the integro-differential or dynamical mean-field equations, which is computationally expensive and has no closed form solutions in general. From the aspect of associated Fokker-Planck equation, the steady state solution is generally unknown. Here, we treat searching for the steady states as an optimization problem, and construct an approximate potential related to the speed of the dynamics, and find that searching for the ground state of this potential is equivalent to running an approximate stochastic gradient dynamics or Langevin dynamics. Only in the zero temperature limit, the distribution of the original steady states can be achieved. The resultant stationary state of the dynamics follows exactly the canonical Boltzmann measure. Within this framework, the quenched disorder intrinsic in the neural networks can be averaged out by applying the replica method, which leads naturally to order parameters for the non-equilibrium steady states. Our theory reproduces the well-known result of edge-of-chaos, and further the order parameters characterizing the continuous transition are derived, and the order parameters are explained as fluctuations and responses of the steady states. Our method thus opens the door to analytically study the steady state landscape of the deterministic or stochastic high dimensional dynamics.

21 pages, 9 figures, revised version 2

References in corpus (18)

• Stochastic thermodynamics, fluctuation theorems, and molecular machines
• The large deviation approach to statistical mechanics
• Fluctuation theorem for stochastic dynamics
• Steady State Thermodynamics of Langevin Systems
• Stochastic Dynamical Structure (SDS) of Nonequilibrium Processes in the Absence of Detailed Balance. II: construction of SDS with nonlinear force and multiplicative noise
• Topological and Dynamical Complexity of Random Neural Networks
• Nonlinear Analogue of the May-Wigner Instability Transition
• Optimal sequence memory in driven random networks
• Correlations between synapses in pairs of neurons slow down dynamics in randomly connected neural networks
• Path integral methods for the dynamics of stochastic and disordered systems
• Origin of the computational hardness for learning with binary synapses
• Dimension of activity in random neural networks
• Counting equilibria in a random non-gradient dynamics with heterogeneous relaxation rates
• Introduction to dynamical mean-field theory of randomly connected neural networks with bidirectionally correlated couplings
• Spectrum of non-Hermitian deep-Hebbian neural networks
• Eight challenges in developing theory of intelligence
• Statistical Mechanics of Recurrent Neural Networks I. Statics
• Random Linear Systems with Quadratic Constraints: from Random Matrix Theory to replicas and back

Cited by in corpus (5)

• Freezing chaos without synaptic plasticity
• Discontinuous phase transition of feature detection in lateral predictive coding
• Synaptic plasticity alters the nature of chaos transition in neural networks
• Network reconstruction may not mean dynamics prediction
• Response function as a quantitative measure of consciousness in brain dynamics