Post Reinforcement Learning Inference

Vasilis Syrgkanis, Ruohan Zhan

We study estimation and inference using data collected by reinforcement learning (RL) algorithms. These algorithms adaptively experiment by interacting with individual units over m…