Self-Evolutionary Reservoir Computer Based on Kuramoto Model

TL;DR

This paper introduces a biologically inspired reservoir computing model that dynamically adapts its structure through co-evolving phase oscillators, enabling autonomous problem-specific adaptation without human intervention.

Contribution

It presents a novel adaptive reservoir model based on phase oscillators that self-evolves its structure in response to inputs, mimicking synaptic plasticity.

Findings

Reservoir structure adapts autonomously to different tasks.

Dynamic co-evolution improves processing of spatiotemporal data.

Model mimics biological synaptic plasticity mechanisms.

Abstract

The human brain's synapses have remarkable activity-dependent plasticity, where the connectivity patterns of neurons change dramatically, relying on neuronal activities. As a biologically inspired neural network, reservoir computing (RC) has unique advantages in processing spatiotemporal information. However, typical reservoir architectures only take static random networks into account or consider the dynamics of neurons and connectivity separately. In this paper, we propose a structural autonomous development reservoir computing model (sad-RC), which structure can adapt to the specific problem at hand without any human expert knowledge. Specifically, we implement the reservoir by adaptive networks of phase oscillators, a commonly used model for synaptic plasticity in biological neural networks. In this co-evolving dynamic system, the dynamics of nodes and coupling weights in the reservoir constantly interact and evolve together when disturbed by external inputs.