2 ms·Compressing Deep Reinforcement Learning Policies into Symbolic Expressions [pdf]3 points by legothief 5y ago