Addressing Limited Weight Resolution in a Fully Optical Neuromorphic Reservoir Computing Readout

TL;DR

This paper introduces a method to enhance the performance of optical neuromorphic reservoir computing readouts with low-resolution weights, achieving near full-resolution accuracy despite noise and limited levels.

Contribution

The paper proposes a novel technique to significantly improve low-resolution optical weights, enabling high-performance neuromorphic computing with fewer resolution levels.

Findings

Outperforms traditional low-resolution weighting by several orders of magnitude.

Achieves performance close to full-resolution weighting in noisy environments.

Effective with only 8 to 32 resolution levels.

Abstract

Using optical hardware for neuromorphic computing has become more and more popular recently due to its efficient high-speed data processing capabilities and low power consumption. However, there are still some remaining obstacles to realizing the vision of a completely optical neuromorphic computer. One of them is that, depending on the technology used, optical weighting elements may not share the same resolution as in the electrical domain. Moreover, noise and drift are important considerations as well. In this article, we investigate a new method for improving the performance of optical weighting, even in the presence of noise and in the case of very low resolution. Even with only 8 to 32 levels of resolution, the method can outperform the naive traditional low-resolution weighting by several orders of magnitude in terms of bit error rate and can deliver performance very close to full-resolution weighting elements, also in noisy environments.