Abstract: How to learn good latent representations is an important topic in the modern era of machine learning. For reinforcement learning, using a good representation makes the decision-making process much more efficient. In this talk, I will cover our work that constructs task-specific latent action space for search-based optimization of black-box functions, finds a representation for policy change that enables joint policy search in imperfect information collaborative games and how different representations affect RL exploration.

Bio: Yuandong Tian is a Research Scientist and Manager in Facebook AI Research, working on deep reinforcement learning and representation learning. He is the lead scientist and engineer for ELF OpenGo and DarkForest Go projects. Prior to that, he was in Google Self-driving Car team in 2013-2014. He received a Ph.D in Robotics Institute, Carnegie Mellon University in 2013. He is the recipient of 2013 ICCV Marr Prize Honorable Mentions.