Neuro-Fuzzy Computing System with the Capacity of Implementation on Memristor-Crossbar and Optimization-Free Hardware Training

TL;DR

This paper introduces a neuro-fuzzy computing system that can be implemented on memristor-crossbar hardware, trained with Hebbian learning, and effectively handles large datasets without overtraining.

Contribution

It presents a novel neuro-fuzzy system with direct hardware implementation and optimization-free training, bridging logical circuits, neural networks, and fuzzy logic.

Findings

Effective implementation on memristor-crossbar hardware

Hardware training using Hebbian learning without optimization

Capable of handling large datasets without overtraining

Abstract

In this paper, first we present a new explanation for the relation between logical circuits and artificial neural networks, logical circuits and fuzzy logic, and artificial neural networks and fuzzy inference systems. Then, based on these results, we propose a new neuro-fuzzy computing system which can effectively be implemented on the memristor-crossbar structure. One important feature of the proposed system is that its hardware can directly be trained using the Hebbian learning rule and without the need to any optimization. The system also has a very good capability to deal with huge number of input-out training data without facing problems like overtraining.