IEEE CIS Seminar: Evolving Players for the Generalized Divide the Dollar Game / Prof. Garrison Greenwood

#Optimization; #Evolutionary #Algorithm; #Neural #network; #CMA-ES; #Bargaining #problem; #Divide #the #dollar;
Share

IEEE Computational Intelligence Society, Australian Capital Territory (ACT) chapter invites you to: 

Seminar (in-person): 07 September 2022, 3.00-4.00 pm AEST (Canberra time)

Speaker: Professor Garrison Greenwood, Portland State University, USA

Title: Evolving Players for the Generalized Divide the Dollar Game

Venue: LTN 08, Building 32 (Lecture Theatre North), UNSW Canberra, Northcott Drive, ACT

===

Abstract:  

Divide the dollar is a simpler version of a game invented by John Nash to study the bargaining problem. Two players simultaneously state (bid) how much of a dollar they will accept. If the bid total is less than or equal to $1, they get their bid as a payoff. Otherwise, both players get nothing. The generalized divide the dollar game is an n-player version that will also, under limited circumstances, allow for small subsidies. In this talk I describe how players for an n=3 player game, implemented as neural networks, can be co-evolved. Both Differential Evolution and CMA-ES versions of the co-evolution are discussed along with an analysis on why one evolutionary algorithm achieved better results.  But co-evolving neural network players, although highly successful, becomes computationally expensive as n grows. A second evolutionary algorithm approach, using a completely different genome, is discussed and results are presented. This second method can successfully evolve 20 or more players with small computational effort.

===

Speaker biography: 

Garrison Greenwood received a Ph.D. in Electrical Engineering from the University of Washington, Seattle, WA. After spending more than a decade in industry designing multiprocessor embedded systems, he entered academia. He is currently a professor in the Electrical and Computer Engineering Department at Portland State University, Portland, OR. Dr. Greenwood has a long history of support of the IEEE Computational Intelligence Society including serving as the general chair of the IEEE Congress on Evolutionary Computation conference (2004 and 2012) and 4 years as the CIS Vice President of Conferences. From 2009 through 2014 he was the Editor-in-Chief of the IEEE Transactions on Evolutionary Computation. His research interests are evolvable hardware and mathematical game theory. Dr. Greenwood is a registered professional engineer in the State of California, USA.

============

 



  Date and Time

  Location

  Hosts

  Registration



  • Add_To_Calendar_icon Add Event to Calendar
  • LTN 08
  • UNSW Canberra, Northcott Drive
  • Canberra, Australian Capital Territory
  • Australia 2602
  • Building: Lecture Theatre North (Building 32)
  • Click here for Map

  • Contact Event Host