Can machine learning for quantum-gas experiments be explainable?
For experimental quantum-gas physicists, this work provides a practical demonstration that interpretable ML models can be effective, addressing the need for explainability in scientific applications.
The authors apply machine learning to denoise raw images and identify solitonic waves in Bose-Einstein condensates, demonstrating that simpler models can achieve comparable performance to complex ones while offering better interpretability.
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.