Neuromorphic computing with multi-memristive synapses

TL;DR

This paper introduces a multi-memristive synaptic architecture with a global arbitration scheme, demonstrating its effectiveness through simulations and experiments with phase change memory devices for scalable, energy-efficient neuromorphic computing.

Contribution

It proposes a novel multi-memristive synapse design with an arbitration scheme and validates it through extensive simulations and large-scale experiments.

Findings

Effective for both spiking and non-spiking neural networks

Successful unsupervised learning with over a million devices

Significant step towards large-scale neuromorphic systems

Abstract

Neuromorphic computing has emerged as a promising avenue towards building the next generation of intelligent computing systems. It has been proposed that memristive devices, which exhibit history-dependent conductivity modulation, could efficiently represent the synaptic weights in artificial neural networks. However, precise modulation of the device conductance over a wide dynamic range, necessary to maintain high network accuracy, is proving to be challenging. To address this, we present a multi-memristive synaptic architecture with an efficient global counter-based arbitration scheme. We focus on phase change memory devices, develop a comprehensive model and demonstrate via simulations the effectiveness of the concept for both spiking and non-spiking neural networks. Moreover, we present experimental results involving over a million phase change memory devices for unsupervised learning of temporal correlations using a spiking neural network. The work presents a significant step towards the realization of large-scale and energy-efficient neuromorphic computing systems.