Can machine learning for quantum-gas experiments be explainable?

TL;DR

This paper explores the potential for explainability in machine learning applications to quantum-gas experiments, focusing on image denoising and soliton identification in Bose-Einstein condensates.

Contribution

It demonstrates ML techniques for image processing in quantum simulators and discusses the balance between model performance, complexity, and interpretability.

Findings

ML effectively denoises raw quantum-gas images

ML identifies solitonic waves in Bose-Einstein condensates

Performance and interpretability trade-offs are analyzed

Abstract

Virtually all aspects of many-body atomic physics are challenging: experiments are technically demanding, datasets have become enormous, and the memory and CPU requirements for classical simulation of generic quantum systems often scale exponentially with system size. Machine learning (ML) methods are already assisting in each of these areas and are poised to become transformative. Here, we focus on two specific applications of ML to cold-atom-based quantum simulators. These devices generally generate data in the form of images; we first showcase denoising of raw images and then identify solitonic waves in Bose-Einstein condensates. In both of these examples, we comment on the interplay between performance, model complexity, and interpretability.