[Invited Talk] Minimizing the reality-gap between simulations and real-world

Presenter: Sungjoon Choi, Disney Research; Time: 4:00pm, Tuesday (2019/07/24); Location: Building 133 Room 204

Abstract

 The landscape of Robotics has been constantly changing over the past decades. Recently, due to the rapid advances in Machine Learning combined with progress in sensing and computing technologies, there have been a number of significant improvements in Robotics including reinforcement learning (RL) and imitation learning (IL). In this talk, I will introduce a number of recent studies regarding modeling latent spaces for efficient learning processes and sim-to-real approaches to minimize the reality-gap between simulations and real-world deployments.

Biography

Sungjoon Choi received Ph.D. in Electrical and Computer Engineering from Seoul National University (2018) and B.S degree in Electrical Engineering and Computer Science from Seoul National University (2012). He is currently a postdoc at Disney Research Los Angeles. Before joining Disney Research, he was a research scientist at Kakao Brain in Korea. His research interests include sample-efficient reinforcement learning and human robot interaction and received Best Conference Paper Finalist Award at 2016 IEEE International Conference on Robotics and Automation (ICRA).