Reservoir Computation with Networks of Differentiating Neuron Ring Oscillators

TL;DR

This paper introduces a novel reservoir computing approach using networks of differentiating neurons in a small world graph, achieving competitive digit recognition performance with potentially lower power consumption.

Contribution

It presents a new reservoir computing model based on differentiating neurons and small world topology, offering an energy-efficient alternative to traditional integrating neuron networks.

Findings

Achieved 90.65% accuracy on MNIST digit recognition.

Demonstrated the viability of differentiating neurons as a reservoir.

Identified network parameters for effective reservoir function.

Abstract

Reservoir Computing is a machine learning approach that uses the rich repertoire of complex system dynamics for function approximation. Current approaches to reservoir computing use a network of coupled integrating neurons that require a steady current to maintain activity. Here, we introduce a small world graph of differentiating neurons that are active only when there are changes in input as an alternative to integrating neurons as a reservoir computing substrate. We find the coupling strength and network topology that enable these small world networks to function as an effective reservoir. We demonstrate the efficacy of these networks in the MNIST digit recognition task, achieving comparable performance of 90.65% to existing reservoir computing approaches. The findings suggest that differentiating neurons can be a potential alternative to integrating neurons and can provide a sustainable future alternative for power-hungry AI applications.