BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//UC Irvine Donald Bren School of Information & Computer Sciences - ECPv6.3.4//NONSGML v1.0//EN
CALSCALE:GREGORIAN
METHOD:PUBLISH
X-WR-CALNAME:UC Irvine Donald Bren School of Information & Computer Sciences
X-ORIGINAL-URL:https://ics.uci.edu
X-WR-CALDESC:Events for UC Irvine Donald Bren School of Information & Computer Sciences
REFRESH-INTERVAL;VALUE=DURATION:PT1H
X-Robots-Tag:noindex
X-PUBLISHED-TTL:PT1H
BEGIN:VTIMEZONE
TZID:America/Los_Angeles
BEGIN:DAYLIGHT
TZOFFSETFROM:-0800
TZOFFSETTO:-0700
TZNAME:PDT
DTSTART:20250309T100000
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:-0700
TZOFFSETTO:-0800
TZNAME:PST
DTSTART:20251102T090000
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTART;TZID=America/Los_Angeles:20250509T110000
DTEND;TZID=America/Los_Angeles:20250509T120000
DTSTAMP:20260307T161213
CREATED:20250430T162320Z
LAST-MODIFIED:20250430T162320Z
UID:25013-1746788400-1746792000@ics.uci.edu
SUMMARY:Learning and Games: A New Frontier in Algorithmic Game Theory
DESCRIPTION:Abstract: Over the past decade\, machine learning has achieved remarkable progress across a wide range of applications. This success is largely attributable to the paradigm of training machine learning systems by minimizing a single loss function using gradient-descent-based optimization algorithms. However\, the landscape is shifting\, and many of the outstanding challenges in machine learning now lie at its intersection with game theory. These challenges arise either from explicitly strategic environments—such as multi-agent reinforcement learning systems\, autobidding in auctions\, and algorithmic pricing—or from problems that can be implicitly modeled as games\, such as robustifying models against adversarial attacks\, training generative models\, and alignment. \nThese emerging problems present new challenges for both machine learning and game theory. On the one hand\, it remains unclear what types of optimization algorithms are effective in these game-theoretic settings. On the other hand\, the games arising from machine learning problems often involve non-concave utilities\, whereas traditional game theory has largely focused on games with (quasi)-concave utilities. In this talk\, we will discuss recent progress in tackling these challenges and explore the obstacles and opportunities that lie ahead for both machine learning and game theory. \nBio: Yang Cai is an Associate Professor of Computer Science and Economics (secondary appointment) at Yale University. He finished his Ph.D. at MIT in Computer Science under the supervision of Costis Daskalakis and received his B.Sc. in EECS at Peking University. His research interests lie in theoretical computer science and its interface with economics\, optimization\, and machine learning. He has been honored with the FOCS Test of Time Award\, the Sloan Research Fellowship\, the NSF CAREER Award\, the William Dawson Scholarship\, and the Simons-Berkeley Research Fellowship. His dissertation has been recognized by the George M. Sprowls Award (for best MIT doctoral theses in CS) and the SIGecom Doctoral Dissertation Award.
URL:https://ics.uci.edu/event/learning-and-games-a-new-frontier-in-algorithmic-game-theory/
LOCATION:Donald Bren Hall\, Irvine\, CA\, 92697\, United States
ATTACH;FMTTYPE=image/jpeg:https://ics.uci.edu/wp-content/uploads/2025/04/Square_Yang-Cai.png_resized.jpg
END:VEVENT
END:VCALENDAR