Events login

Division of Research | Research Events

Warning Icon This event is in the past.
January 16, 2018 | 11:30 a.m. - 12:20 p.m.
Category: Seminar
Location: State Hall #101 | Map
5143 Cass
Detroit, MI 48202
Cost: Free
Audience: Academic Staff, Alumni, Community, Current Graduate Students, Current Undergraduate Students, Faculty, Parents, Prospective Students, Staff

Game Theoretic Knowledge Distribution in Cultural Algorithms


Cultural Algorithms (CA) is a socially motivated, hyperheuristic approach for finding solutions in a search landscape. The CA employs a high level component called the Belief Space that harvests/disseminates varied types of knowledge from/to the population of individuals. The Belief Space consists of Knowledge Sources (KS), each of which are essentially search strategies.  Some of them are exploratory while others are exploitative. The Cultural Algorithm coordinates the selection and application of these search heuristics during the problem solving process. A Knowledge Distribution mechanism controls how knowledge from the Belief Space is distributed among the population of individuals. Achieving the right balance between the influences of the various types of knowledge on the population, is a key goal of CA. Over the years, many Knowledge Distribution mechanisms have been proposed and studied. They have focused on the selection of just one approach to influence an individual. Here we propose a new mechanism based on game theoretic ideas that supports the cooperation between knowledge sources, .and empirically show that the mechanism performs significantly better than the most commonly used mechanism called Weighted Majority Win (WMW) on a diverse set of problem landscapes.


Dr. Robert G. Reynolds received his Ph.D.  in Computer Science, specializing in Artificial Intelligence from the University of Michigan, Ann Arbor. He is currently a Professor of Computer Science and director of the Artificial Intelligence Laboratory at Wayne State University.  He is a Senior Member of the IEEE. At the University of Michigan-Ann Arbor, Professor Reynolds is a Visiting Research Scientist with the Museum of Anthropology, and a member of the Complex Systems Group. His interests are in the development of computational models of cultural evolution for use in the simulation of complex organizations, computer gaming, and virtual world applications.

Dr. Reynolds produced a framework called Cultural Algorithms, to express and computationally test various theories of social evolution using multi-agent simulation models. He has authored or co-authored over 250 papers. Currently, Dr. Reynolds along with his students, are developing a toolkit for testing Cultural Algorithms in dynamic environments; the Cultural Algorithm Toolkit (CAT). His research group has produced award winning game controller software for several international competitions using the Cultural Algorithms toolkit.  In 2017, a software system based upon Cultural Algorithms came in second in the IEEE Single Real Valued Function Optimization competition held in conjunction with the IEEE Congress on Evolutionary Computation.

Dr. Reynolds has applied Cultural Algorithms to problems in social evolution including the evolution of agriculture; the origins of the state in Ancient Mexico; the discovery of ancient hunting sites underneath Lake Huron; the emergence of prehistoric urban centers in Mexico; the origins of language and culture in Peru; and the disappearance of the Ancient Anazazi in Southwestern Colorado. He has co-authored three books in this area that include the following; Flocks of the Wamani (1989, Academic Press), with Joyce Marcus and Kent V. Flannery; The Acquisition of Software Engineering Knowledge (2003, Academic Press), with George Cowan; and Excavations at San Jose Mogote 1: The Household Archaeology with Kent Flannery and Joyce Marcus (2005, Museum of Anthropology-University of Michigan Press). Another book entitled Culture on the Edge of Chaos: Cultural Algorithms and the Foundations of Social Intelligence is scheduled to appear in 2018. 

For more information about this event, please contact LaNita Stewart at 313-577-2478 or