Beyond catastrophic forgetting in associative networks with self-interactions

Abstract

Spin-glass models of associative memories are a cornerstone between statistical physics and theoretical neuroscience. In these networks, stochastic spin-like units interact through a synaptic matrix shaped by local Hebbian learning. In absence of self-interactions (i.e., autapses), the free energy reveals catastrophic forgetting of all stored patterns when their number exceeds a critical memory load. Here, we bridge the gap with biology by considering networks of deterministic, graded units coupled via the same Amari-Hopfield synaptic matrix, while retaining autapses. Contrary to the assumption that self-couplings play a negligible role, we demonstrate that they qualitatively reshape the energy landscape, confining the recurrent dynamics to the subspace hosting the stored patterns. This allows for the derivation of an exact overlap-dependent Lyapunov function, valid even for networks with finite size. Moreover, self-interactions generate an auxiliary internal field aligned with the target memory pattern, widening the repertoire of accessible attractor states. Consequently, pure recall states act as robust associative memories for any memory load, beyond the critical threshold for catastrophic forgetting observed in spin-glass models -- all without requiring nonlocal learning prescriptions or significant reshaping of the Hebbian synaptic matrix.