Symmetry-Aware Reservoir Computing

TL;DR

This paper introduces a symmetry-aware reservoir computing approach that significantly enhances processing power and efficiency for tasks with known symmetries, outperforming traditional methods in parity and chaotic system inference tasks.

Contribution

The authors develop a method to incorporate symmetry properties into reservoir computing, reducing data and network size requirements while improving accuracy on symmetry-dependent tasks.

Findings

Zero error achieved with exponentially reduced data and network size.

Linear scaling of network size with parity order for zero error.

Order of magnitude improvement in inference accuracy over regular RCs.

Abstract

We demonstrate that matching the symmetry properties of a reservoir computer (RC) to the data being processed dramatically increases its processing power. We apply our method to the parity task, a challenging benchmark problem that highlights inversion and permutation symmetries, and to a chaotic system inference task that presents an inversion symmetry rule. For the parity task, our symmetry-aware RC obtains zero error using an exponentially reduced neural network and training data, greatly speeding up the time to result and outperforming hand crafted artificial neural networks. When both symmetries are respected, we find that the network size $N$ necessary to obtain zero error for 50 different RC instances scales linearly with the parity-order $n$. Moreover, some symmetry-aware RC instances perform a zero error classification with only $N=1$ for $n\leq7$. Furthermore, we show that a symmetry-aware RC only needs a training data set with size on the order of $(n+n/2)$ to obtain such performance, an exponential reduction in comparison to a regular RC which requires a training data set with size on the order of $n2^n$ to contain all $2^n$ possible $n-$bit-long sequences. For the inference task, we show that a symmetry-aware RC presents a normalized root-mean-square error three orders-of-magnitude smaller than regular RCs. For both tasks, our RC approach respects the symmetries by adjusting only the input and the output layers, and not by problem-based modifications to the neural network. We anticipate that generalizations of our procedure can be applied in information processing for problems with known symmetries.