Performance of machine-learning-assisted Monte Carlo in sampling from simple statistical physics models.

Recent years have seen a rise in the application of machine learning techniques to aid the simulation of hard-to-sample systems that cannot be studied using traditional methods. Despite the introduction of many different architectures and procedures,…

Preview only. Read the full abstract at the source