Combining machine learning with physics: A framework for tracking and sorting multiple dark solitons

Abstract

In ultracold-atom experiments, data often comes in the form of images which suffer information loss inherent in the techniques used to prepare and measure the system. This is particularly problematic when the processes of interest are complicated, such as interactions among excitations in Bose-Einstein condensates (BECs). In this paper, we describe a framework combining machine learning (ML) models with physics-based traditional analyses to identify and track multiple solitonic excitations in images of BECs. We use an ML-based object detector to locate the solitonic excitations and develop a physics-informed classifier to sort solitonic excitations into physically motivated subcategories. Lastly, we introduce a quality metric quantifying the likelihood that a specific feature is a longitudinal soliton. Our trained implementation of this framework, SolDet, is publicly available as an open-source python package. SolDet is broadly applicable to feature identification in cold-atom images when trained on a suitable user-provided dataset.

13 pages, 9 figures

References in corpus (9)

• Quantum phase transition from a superfluid to a Mott insulator in a gas of ultracold atoms
• Spontaneous symmetry breaking in a quenched ferromagnetic spinor Bose condensate
• Oscillations and interactions of dark and dark-bright solitons in Bose-Einstein condensates
• A Review on Generative Adversarial Networks: Algorithms, Theory, and Applications
• Observation of Solitonic Vortices in Bose-Einstein Condensates
• Stabilization of Solitons Generated by a Supersonic Flow of Bose-Einstein Condensate Past an Obstacle
• Linear "ship waves" generated in stationary flow of a Bose-Einstein condensate past an obstacle
• Creating solitons with controllable and near zero velocity in Bose-Einstein condensates
• Experimentally Realizing Efficient Quantum Control with Reinforcement Learning

Cited by in corpus (2)

• Vortex detection in atomic Bose-Einstein condensates using neural networks trained on synthetic images
• Dark solitons in Bose-Einstein condensates: a dataset for many-body physics research