I’m broadly interested in the intersection of reinforcement learning and language. In particular, I’m looking at alternatives to perform human-in-the-loop learning with minimal human supervision.