# Wave Physics as an Analog Recurrent Neural Network

## TL;DR

This paper demonstrates that wave physics systems can be mapped to recurrent neural networks, enabling analog hardware to perform complex temporal data processing efficiently, exemplified by vowel classification with comparable accuracy to digital methods.

## Contribution

It introduces a novel mapping between wave physics dynamics and RNN computation, enabling training of physical wave systems for machine learning tasks.

## Key findings

- Wave physics can be used to implement RNN-like computation.
- An inhomogeneous medium can classify vowels from raw audio.
- Performance is comparable to digital RNNs.

## Abstract

Analog machine learning hardware platforms promise to be faster and more energy-efficient than their digital counterparts. Wave physics, as found in acoustics and optics, is a natural candidate for building analog processors for time-varying signals. Here we identify a mapping between the dynamics of wave physics, and the computation in recurrent neural networks. This mapping indicates that physical wave systems can be trained to learn complex features in temporal data, using standard training techniques for neural networks. As a demonstration, we show that an inverse-designed inhomogeneous medium can perform vowel classification on raw audio signals as their waveforms scatter and propagate through it, achieving performance comparable to a standard digital implementation of a recurrent neural network. These findings pave the way for a new class of analog machine learning platforms, capable of fast and efficient processing of information in its native domain.