Mean Field Theory of Dynamical Systems Driven by External Signals

Abstract

Dynamical systems driven by strong external signals are ubiquituous in nature and engineering. Here we study "echo state networks", networks of a large number of randomly connected nodes, which represent a simple model of a neural network, and have important applications in machine learning. We develop a mean field theory of echo state networks. The dynamics of the network is captured by the evolution law, similar to a logistic map, for a single collective variable. When the network is driven by many independent external signals, this collective variable reaches a steady state. But when the network is driven by a single external signal, the collective variable is nonstationnary but can be characterised by its time averaged distribution. The predictions of the mean field theory, including the value of the largest Lyaponuov exponent, are compared with the numerical integration of the equations of motion.

7 pages, 6 figures

References in corpus (2)

• Optoelectronic Reservoir Computing
• Noise-induced behaviors in neural mean field dynamics

Cited by in corpus (17)

• Local Dynamics in Trained Recurrent Neural Networks
• Deep Echo State Network (DeepESN): A Brief Survey
• Edge of chaos and avalanches in neural networks with heavy-tailed synaptic weight distribution
• Determination of the edge of criticality in echo state networks through Fisher information maximization
• Consistency in Echo-State Networks
• Optimal short-term memory before the edge of chaos in driven random recurrent networks
• Learn to Synchronize, Synchronize to Learn
• Quality of internal representation shapes learning performance in feedback neural networks
• Suppression of chaos in a partially driven recurrent neural network
• Local homeostatic regulation of the spectral radius of echo-state networks
• Neuronal correlations shape the scaling behavior of memory capacity and nonlinear computational capability of reservoir recurrent neural networks
• Convolutional Reservoir Computing for World Models
• Phase transitions from linear to nonlinear information processing in neural networks
• Reinforcement Learning with Convolutional Reservoir Computing
• Speech Emotion Recognition System by Quaternion Nonlinear Echo State Network
• A geometrical analysis of global stability in trained feedback networks
• Memory Uncertainty Relation and Harmonic Memory in Random Recurrent Networks