Chaos-guided Input Structuring for Improved Learning in Recurrent Neural Networks

TL;DR

This paper demonstrates that structuring inputs along dominant chaotic projections stabilizes neural trajectories in recurrent networks, enhancing their computational and inference capabilities by controlling attractor stability.

Contribution

It introduces a novel input structuring method based on phase alignment with chaotic activity to improve RNN performance, supported by mean field analysis.

Findings

Input structuring stabilizes chaotic trajectories.

Aligned inputs enhance noise suppression and attractor stability.

Improved inference in recurrent neural networks.

Abstract

Anatomical studies demonstrate that brain reformats input information to generate reliable responses for performing computations. However, it remains unclear how neural circuits encode complex spatio-temporal patterns. We show that neural dynamics are strongly influenced by the phase alignment between the input and the spontaneous chaotic activity. Input structuring along the dominant chaotic projections causes the chaotic trajectories to become stable channels (or attractors), hence, improving the computational capability of a recurrent network. Using mean field analysis, we derive the impact of input structuring on the overall stability of attractors formed. Our results indicate that input alignment determines the extent of intrinsic noise suppression and hence, alters the attractor state stability, thereby controlling the network's inference ability.