
Abstract: Reinforcement Learning (RL) is a technique for solving sequential decision-making problems that achieved superhuman performance in games such as Chess and GO. In recent years, RL algorithms have matured and are readily been applied in diverse contexts such as robotics, urban planning, healthcare, logistics, recommendation systems, etc. However, successfully training RL systems remains an art, and several considerations govern if RL is the right tool for a new problem. In this talk, we will discuss these considerations, overview some successful applications, and highlight major open issues and some dirty secrets behind the practice of RL.
Bio: Pulkit is an Assistant Professor in the department of Electrical Engineering and Computer Science (EECS) at MIT. His lab is a part of the Computer Science and Artificial Intelligence Lab (CSAIL), is affiliated with the Laboratory for Information and Decision Systems (LIDS) and involved with NSF AI Institute for Artificial Intelligence and Fundamental Interactions ( IAIFI ).
He completed Ph.D. at UC Berkeley; undergraduate studies from IIT Kanpur. Co-founded SafelyYou Inc. that builds fall prevention technology.