Applying machine learning techniques to ultracold quantum gases

Machine Learning (ML) techniques have emerged as a powerful tool in quantum matter research owing to its ability in analyzing large datasets. Recently, various quantum systems have been explored by applying ML algorithms to data both from numerical simulations and from experiments. In this talk, we demonstrate the benefit of ML technique in quantum gas experiments including the thermodynamic measurement of SU(N) fermions, the detection of topological phase transition out of spin texture, and the thermometry of a Fermi gas. We find that ML-aided analysis efficiently guides us to investigate the useful information and facilitate research in quantum gas systems. Our works complement recent ML studies of quantum many-body physics to explore the underlying physics.