Hebbian Attractor Networks for Robot Locomotion

TL;DR

This paper introduces Hebbian Attractor Networks (HAN), which utilize local Hebbian plasticity with dual-timescale updates to enable adaptive control in robotic locomotion, demonstrating how timing influences neural dynamics and stability.

Contribution

The paper presents a novel class of plastic neural networks, HAN, that induce attractor dynamics through local weight normalization and dual-timescale Hebbian updates, advancing adaptive control in robotics.

Findings

Slower updates lead to stable weight convergence.

Faster updates produce oscillatory attractor dynamics.

Results generalize to high-dimensional quadrupedal robots.

Abstract

Biological neural networks continuously adapt and modify themselves in response to experiences throughout their lifetime - a capability largely absent in artificial neural networks. Hebbian plasticity offers a promising path toward rapid adaptation in changing environments. Here, we introduce Hebbian Attractor Networks (HAN), a class of plastic neural networks in which local weight update normalization induces emergent attractor dynamics. Unlike prior approaches, HANs employ dual-timescale plasticity and temporal averaging of pre- and postsynaptic activations to induce either co-dynamic limit cycles or fixed-point weight attractors. Using simulated locomotion benchmarks, we gain insight into how Hebbian update frequency and activation averaging influence weight dynamics and control performance. Our results show that slower updates, combined with averaged pre- and postsynaptic activations, promote convergence to stable weight configurations, while faster updates yield oscillatory co-dynamic systems. We further demonstrate that these findings generalize to high-dimensional quadrupedal locomotion with a simulated Unitree Go1 robot. These results highlight how the timing of plasticity shapes neural dynamics in embodied systems, providing a principled characterization of the attractor regimes that emerge in self-modifying networks.