Optimal sequence memory in driven random networks

2016/03/31 by Jannis Schuecker, Sven Goedeke, Moritz Helias · 2 citations
Biochemistry, Genetics and Molecular Biology · Physics and Astronomy · #q-bio.NC #nlin.CD

paper · pdf · doi:10.1103/physrevx.8.041029

published as Phys. Rev. X 8, 041029 (2018)

arxiv created 2017/09/22 · arxiv updated 2018/11/21

Abstract

Autonomous randomly coupled neural networks display a transition to chaos at a critical coupling strength. We here investigate the effect of a time-varying input on the onset of chaos and the resulting consequences for information processing. Dynamic mean-field theory yields the statistics of the activity, the maximum Lyapunov exponent, and the memory capacity of the network. We find an exact condition that determines the transition from stable to chaotic dynamics and the sequential memory capacity in closed form. The input suppresses chaos by a dynamic mechanism, shifting the transition to significantly larger coupling strengths than predicted by local stability analysis. Beyond linear stability, a regime of coexistent locally expansive, but non-chaotic dynamics emerges that optimizes the capacity of the network to store sequential input.

Cited by

• Random neural networks match observed dimensionality of neural population recordings and motivate stronger experimental tests
• Task-specific programming of chaos in neural circuits

Related

• Transition to chaos in random networks with cell-type-specific connectivity
• Solvable Neural Network Model for Input-Output Associations: Optimal Recall at the Onset of Chaos
• Phase transitions from linear to nonlinear information processing in neural networks
• Contraction, Criticality, and Capacity: A Dynamical-Systems Perspective on Echo-State Networks
• Topological and Dynamical Complexity of Random Neural Networks