Bradly Stadie received a PhD from UC Berkeley in 2018. Prior to that, he received an BA from the University of Chicago in 2014. Bradly spent two years as a research scientist at OpenAI from 2016-2017.
His research is in the field of reinforcement learning. In particular, his work on handling distributional shifts in sparse reward and imitation settings has improved the generalization abilities of robots both in simulation and the real world. This line of work also provides a good platform for better understanding causal inference in the context of high dimensional auto-correlated time series.
Dr. Stadie also has a personally maintained website which can be found at http://www.ttic.edu/stadie.