Deep Photonic Reservoir Computer Based on Frequency Multiplexing with Fully Analog Connection Between Layers

TL;DR

This paper introduces a fully analog, fiber-based deep photonic reservoir computer using frequency multiplexing, demonstrating significant performance improvements over traditional RCs in processing time series tasks.

Contribution

It presents a novel two-layer deep photonic reservoir computer with frequency multiplexing and fully analog inter-layer connections, enhancing performance without digital processing.

Findings

Deep-RC outperforms traditional RC by up to 100x on benchmark tasks.

Fully analog, fiber-based implementation reduces digital conversion costs.

Demonstrates potential for complex time series processing in photonic neuromorphic systems.

Abstract

Reservoir computers (RC) are randomized recurrent neural networks well adapted to process time series, performing tasks such as nonlinear distortion compensation or prediction of chaotic dynamics. Deep reservoir computers (deep-RC), in which the output of one reservoir is used as the input for another one, can lead to improved performance because, as in other deep artificial neural networks, the successive layers represent the data in more and more abstract ways. We present a fiber-based photonic implementation of a two-layer deep-RC based on frequency multiplexing. The two RC layers are encoded in two frequency combs propagating in the same experimental setup. The connection between the layers is fully analog and does not require any digital processing. We find that the deep-RC outperforms a traditional RC by up to two orders of magnitude on two benchmark tasks. This work paves the way towards using fully analog photonic neuromorphic computing for complex processing of time series, while avoiding costly analog-to-digital and digital-to-analog conversions.