Hier finden Sie wissenschaftliche Publikationen aus den Fraunhofer-Instituten.

Optimal Sequence Memory in Driven Random Networks

: Schücker, Jannis; Goedeke, Sven; Helias, Moritz

Fulltext urn:nbn:de:0011-n-5523597 (1.7 MByte PDF)
MD5 Fingerprint: 27e74631de3b7bf7748ab8f667b1e7c1
(CC) by
Created on: 26.7.2019

Physical review. X, Expanding access 8 (2018), No.4, Art. 041029, 28 pp.
ISSN: 2160-3308
Journal Article, Electronic Publication
Fraunhofer IAIS ()
neuronal network; chaotic system; stochastic dynamical system; mean field theory; path-integral method; Nonlinear Dynamics; statistical physics; Networks; biological physics; interdisciplinary physics

Autonomous, randomly coupled, neural networks display a transition to chaos at a critical coupling strength. Here, we 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 nonchaotic dynamics emerges that optimizes the capacity of the network to store sequential input.