BEGIN:VCALENDAR
PRODID:-//eluceo/ical//2.0/EN
VERSION:2.0
CALSCALE:GREGORIAN
BEGIN:VEVENT
UID:e19e6b151386c77e4c11ef70ea7d8194
DTSTAMP:20260407T103638Z
SUMMARY:QUADS Seminar: Payment Rules through Discriminant-Based Classifiers
  (David Parkes\, Harvard)
DESCRIPTION:Title: Payment Rules through Discriminant-Based ClassifiersJoin
 t work with P. Dutting\, F. Fischer\, P. Jiarapniyo\, J. K. Lai\, and B. L
 ubin http://www.seas.harvard.edu/directory/parkes\nAbstract: In mechanism 
 design it is typical to impose incentive compatibility and then derive an 
 optimal mechanism subject to this constraint. By replacing the incentive c
 ompatibility requirement with the goal of minimizing expected ex-post regr
 et\, we are able to adapt techniques of statistical machine learning to th
 e design of payment rules. This computational approach to mechanism design
  is applicable to domains with multi- dimensional types and situations whe
 re computational efficiency is a concern. Specifically\, given an outcome 
 rule and access to a type distribution\, we train a support vector machine
  with a special discriminant function structure such that it implicitly es
 tablishes a payment rule with desirable incentive properties. We discuss a
 pplications to a multi-minded combinatorial auction with a greedy winner-d
 etermination algorithm and to an assignment problem with egalitarian outco
 me rule. Experimental results demonstrate both that the construction produ
 ces payment rules with low ex-post regret\, and that penalizing classifica
 tion errors is effective in preventing failures of ex-post individual rati
 onality. \nBio: David C. Parkes is the Gordon McKay Professor of Computer 
 Science in the School of Engineering and Applied Sciences at Harvard Unive
 rsity. He was the recipient of the NSF Career Award\, the Alfred P. Sloan 
 Fellowship\, the Thouron Scholarship\, the Harvard University Roslyn Abram
 son Award for Teaching. Parkes received his Ph.D. degree in Computer and I
 nformation Science from the University of Pennsylvania in 2001\, and an M.
 Eng. (First class) in Engineering and Computing Science from Oxford Univer
 sity in 1995. At Harvard\, Parkes founded the Economics and Computer Scien
 ce research group. His research interests include mechanism design\, elect
 ronic commerce\, market design\, and social computing. Parkes is editor of
  Games and Econonic Behavior\, and on the editorial boards of JAAMAS\, ACM
  TEAC and INFORMS J. Computing. Parkes also serves as the Chair of the ACM
  SIG on Electronic Commerce and was the Program Chair of ACMEC’07 and AA
 MAS’08\, and General Chair of ACMEC’10.
URL:https://www.imperial.ac.uk/events/110657/quads-seminar-payment-rules-th
 rough-discriminant-based-classifiers-david-parkes-harvard/
DTSTART;TZID=Europe/London:20120321T140000
DTEND;TZID=Europe/London:20120321T150000
LOCATION:United Kingdom
END:VEVENT
BEGIN:VTIMEZONE
TZID:Europe/London
BEGIN:STANDARD
DTSTART:20120321T140000
TZNAME:GMT
TZOFFSETTO:+0000
TZOFFSETFROM:+0000
END:STANDARD
END:VTIMEZONE
END:VCALENDAR
