Title: Machine Theory of Behavior and Mind

Abstract: We have seen impressive results from machine learning that enable machines to recognize objects very accurately, translate between multiple languages, and manipulate various objects with a high success rate. However, there is still a gap to build machines that can operate in an unconstrained environment. Humans can understand the goal, intention, and desire of other agents, and even build their understanding of the world based on raw sensory observations. However, this type of ability is still missing from the current intelligence systems. In this talk, I will discuss how we can design learning environments, algorithms, and systems to enable machines to build a physical understanding of the world, understand the policy and goal of other agents, and predict action intentionality, all from a large amount of raw and unlabeled data through active interaction with the environment..

Biography: Boyuan Chen is a Ph.D. candidate in the Department of Computer Science at Columbia University. He obtained his Bachelor’s Degree in Biomedical Engineering from the College of Electronics and Engineering, Jilin University. He was a visiting scholar at the University of Manchester in 2016. He studied with Prof. Hod Lipson, a pioneer in the World’s Robotics, Data Science, and 3D printing. He has published many papers in the top international conferences in the field of Artificial Intelligence and Robotics (e.g., NeurIPS, IROS, GECCO, Humanoids, etc.). He served as a reviewer for several international conferences (e.g., CVPR, ICML, ICLR, PRCV, etc.). He has been awarded the ACM GECCO Student Scholarship, Chinese Government Scholarship, and Chinese Academy of Sciences Space Science and Technology Innovation Scholarship.

Personal page: http://www.cs.columbia.edu/~bchen/

Time: 10: 00am, Wednesday, December 25, 2019

Venue: School of Artificial Intelligence (Room 601, Administration Building, Central Campus)

Organizer: School of Artificial Intelligence, Jilin University