报告题目(Title): Ensemble Genetic Programming for Decision Making in Uncertain Environments
报告人姓名(Speaker): Dr. Yi Mei, Evolutionary Computation Research Group, School of Engineering and Computer Science, Victoria University of Wellington
报告摘要(Abstract): Genetic Programming (GP) has achieved great success in evolving heuristics/rules for complex sequential decision making problems in uncertain/dynamic environments. This talk focuses on combinatorial optimisation problems, such as scheduling and arc routing problems. Unlike pre-optimising a solution which may become infeasible in practice, a heuristic/rule evolved by GP builds the solution in real time based on the up-to-date information about the environment in an reactive way. Currently, an important challenge for GP to evolve heuristics is that the evolved heuristics are too complex and hard to interpret. To address this issue, we propose to use ensemble learning to evolve a group of simple heuristics rather than a single complex heuristic, and considered dynamic job shop scheduling and uncertain arc routing problem as our case studies. By trying different ensemble learning approaches, we have achieved some promising results. This suggests the potential of improving both effectiveness and interpretability using ensemble learning.
Dr. Yi Mei is a Senior Lecturer at the School of Engineering and Computer Science, Victoria University of Wellington, Wellington, New Zealand. His research interests include evolutionary scheduling and combinatorial optimisation, machine learning, genetic programming, and hyper-heuristics. He has more than 100 fully referred publications, including the top journals in EC and Operations Research such as IEEE TEVC, IEEE TCYB, Evolutionary Computation Journal, European Journal of Operational Research, ACM Transactions on Mathematical Software. He received an IEEE Transactions on Evolutionary Computation (top journal in evolutionary computation) Outstanding Paper Award in 2017. He serves as a Vice-Chair of the IEEE CIS Emergent Technologies Technical Committee, and a member of Intelligent Systems Applications Technical Committee. He is an Editorial Board Member of International Journal of Bio-Inspired Computation, and a guest editor of a special issue of the Genetic Programming Evolvable Machine journal. He serves as a reviewer of over 30 international journals. He is an IEEE Senior Member.