FDS Seminar: John Langford (Microsoft Research)
Weekly Seminar
Event time:
Wednesday, December 7, 2022 - 4:00pm
Location:
DL220
10 Hillhouse Avenue
New Haven, CT
Event description:
FDS Seminar
“Discovering an Agent’s Configuration Space”
Speaker: John Langford
Microsoft Research New York
Abstract: Given potentially very rich and redundant sensory information which an agent receives, can the agent discover a configuration space summarizing all the configurations it can achieve in an environment? In 2-d free space this could be x,y,theta, but in environments with walls, movable objects, and other agents how can this be done? I’ll discuss a basic principle, theory, and experimental results suggesting this may be viable for _all_ agents with relatively few environment interactions.
Bio: John Langford is a computer scientist working in machine learning and learning theory, a field that he says “is shifting from an academic discipline to an industrial tool”.
He is well known for work on the Isomap embedding algorithm, CAPTCHA challenges, Cover Trees for nearest neighbor search, Contextual Bandits (which he coined) for reinforcement learning applications, and learning reductions.
John is the author of the blog hunch.net and the principal developer of Vowpal Wabbit. He works at Microsoft Research New York, of which he was one of the founding members, and was previously affiliated with Yahoo! Research, Toyota Technological Institute at Chicago, and IBM’s Watson Research Center. He studied Physics and Computer Science at the California Institute of Technology, earning a double bachelor’s degree in 1997, and he received his Ph.D. in Computer Science from Carnegie Mellon University in the year of 2002.
John was the program co-chair for the 2012 International Conference on Machine Learning (ICML), general chair for the 2016 ICML, and is the President of ICML from 2019–2021.
Website: https://hunch.net/~jl/
This is an in-person seminar, with remote access available: https://yale.hosted.panopto.com/Panopto/Pages/Viewer.aspx?id=1a9628bf-c5c4-4830-948b-af420136ee87
Add event to calendar
.