Site Search
Computer Science

Photos

Dr. Robert KentDr. Robert Kent
Dr. Robert Kent
Dr. Ziad Kobti lecturingDr. Ziad Kobti
Dr. Ziad Kobti
Lambton TowerLambton Tower
Lambton Tower
Windsor WaterfrontWindsor Waterfront Park
Windsor Waterfront Park
Jessica Chen, Ph.D.Dr. Jessica Chen
Dr. Jessica Chen
Dr. Scott GoodwinDr. Scott Goodwin
Dr. Scott Goodwin
Dr. Luis RuedaDr. Luis Rueda
Dr. Luis Rueda
Alioune Ngom, Ph.D.Dr. Alioune Ngom
Dr. Alioune Ngom
Arunita Jaekel, Ph.D.Dr. Arunita Jaekel
Dr. Arunita Jaekel
Christie Ezeife, Ph.D.Dr. Christie Ezeife
Dr. Christie Ezeife
Xiaobu Yuan, Ph.D.Dr. Xiaobu Yuan
Dr. Xiaobu Yuan
Robin Gras, Ph.D.Dr. Robin Gras
Dr. Robin Gras
Imran Ahmad, Ph.D.Dr. Imran Ahmad
Dr. Imran Ahmad

A Multi-Level Cooperative Multi-Population Cultural Algorithm

Add this event into your calendar using the iCAL format
  • Mon, 03/12/2018 - 4:00pm - 5:30pm




A Multi-Level Cooperative Multi-Population Cultural Algorithm

MSc Thesis Proposal by:

Dilpreet Singh

Date:  Monday, March 12th, 2018
Time:  4: 00 pm – 5:30 pm
Location: 3105, Lambton Tower

Abstract:

The aim of the research is to improve the capability of Multi-Population Cultural Algorithm to tackle the evolution of cooperation. We are proposing Multilevel Cooperative Multi-Population Cultural Algorithm (ML-MPCA) based on biological group selection theory proposed by Trauslen to encourage cooperation. Any member of each group can only interact with the members of the same group while each group contains at least one individual. The selection process can operate within and between the groups. The population of individuals is divided into subgroups. Within group selection favors individual with higher fitness values inside group. Individuals therefore, compete within the group. This competition in groups encourages cooperation. Between-group selection selects the group that has higher fitness value which means that it will select the group that cooperates the best. In other words, the evolution here occurs at two levels, individual level and at group level. We evaluate the performance of our algorithm on number of benchmark functions and analyze our framework using n-player prisoner’s dilemma problem, and compare the obtained results with other well-known similar algorithms in the field. We expect to see that our proposed algorithm improves both the accuracy of the identified near optimal solution and its required processing time. The model is extendable to more than two levels of selection and can also support individual and group migration.

Thesis Committee:
Internal Reader: Dr. Mehdi Kargar
External Reader: Dr. Roozbeh Razavi Far
Advisor: Dr. Ziad Kobti
Co-Advisor: Dr. Pooya Moradian Zadeh

 



csgradinfo@uwindsor.ca
(519)253-3000


See More: