NeurIPS 2020

Deep Energy-based Modeling of Discrete-Time Physics

Meta Review

This paper has received 3 very favorable reviews, describing a rigorous and well written paper proposing a new method for data-driven modeling of dynamical systems, an important field of research. The reviewers laud novelty, applicability to a broader class of problems, empirical results on a variety of problems, and excellent positionning wrt the state of the art. The AC concurs.