Multi-parameter optimization of polarization gradient cooling for 87 Rb atoms based on reinforcement learning

Polarization gradient cooling (PGC) plays an important role in many cold atom applications including the formation of Bose-Einstein condensates (BECs) and cooling of single atoms. Traditional parameter optimization of PGC usually relies on subjective expertise, faces challenges in fine manipulation, and exhibits low optimization efficiency. Here, we propose a segmented control method that differs from the traditional PGC process by expanding the experiment parameters from 3 to 30. Subsequently, the conventional timing optimization problem is reformulated as a Markov decision process (MDP), and the experiment parameters are optimized using a reinforcement learning model. With proper settings of hyper parameters, the learning process exhibits good convergence and powerful parameter exploration capabilities. Finally, we capture similar to 4.3 x 108 cold atoms, with a phase space density of similar to 7.1 x 10-4 at a temperature of similar to 3.7 mu K in similar to 18.8 min. Our work paves the way for the intelligent preparation of degenerate quantum gas. (c) 2024 Optica Publishing Group under the terms of the Optica Open Access Publishing Agreement