Evolutionary Algorithms for Numerical Constrained Optimization Problems
Speaker: Ming Yuchi, Griffith University
When: 2006-11-16 13:00:00
Venue: G23 2.07A Gold Coast campus, Griffith University
Host: (seminar host unavailable)
Abstract:In recent years, evolutionary algorithms have received a lot of
attention regarding their potential for solving the numerical
constrained optimization problems. When Evolutionary algorithms are
used for solving constraint optimization problems, how to deal with
the relationship between the feasible and infeasible parents
directly influences the quality of the final results. The ratio of
feasible / infeasible parents in a population is
investigated. Through experimental studies on benchmark and
hypothetical problems, it is revealed that the solutions are very
dependent on how the ratio of feasible / infeasible parents
balanced. Also, this is dependent on the problem itself - what may
be an optimal approach for one problem may not be optimal for
another.
To fully utilize the scope between the upper and lower bounds, an
evolutionary algorithm using feasibility-based grouping is
proposed. Feasible and infeasible individuals are divided into two
groups: feasible group and infeasible group. The evaluation and
ranking of these two groups are performed separately. Experimental
results show the efficiency of the proposed method.
Biography:Ming Yuchi, currently works as a postdoctoral research fellow in
ICT, Griffith University, Gold Coast campus. He got his PhD in Korea
Advanced Institute of Science and Technology (KAIST) on February
2006. His main research interests focus on Artificial Intelligence,
Evolutionary Algorithm, Fuzzy System, Neural Network,
Natural-inspired System and Computation, Optimization and Control
System
Type: Institute for Integrated and Intelligent Systems seminar
Contact:(seminar host unavailable), seminar host (n.dunstan@griffith.edu.au)
or Guido Governatori (ITEE seminar co-ordinator)
(guido@itee.uq.edu.au)