An empirical risk minimization approach for offline inverse RL and Dynamic Discrete Choice models | Best AI papers explained | Podwise