Genetic Programming Hyper-heuristics for Combinatorial Optimisation: Yi Mei CIS Webinar

1020 views
Download
  • Share
Create Account or Sign In to post comments
#Evolutionary algorithms #Genetic programming #Hyperheuristics #Combinatorial optimizsation #Job shop scheduling #Arc routing problem #Traveling thief problem

This presentation will give an introduction on the application of GP in evolving heuristics for combinatorial optimisation problems, which are usually important due to their wide real-world applications.

Genetic Programming (GP) has been successfully used as a hyper-heuristic in many areas, particularly for handling dynamic environment and scaling up to huge problem sizes. This presentation will give an introduction on the application of GP in evolving heuristics for combinatorial optimisation problems, which are usually important due to their wide real-world applications. The key difference between hyper-heuristic and traditional optimisation methods is that traditional optimisation methods searches for solutions for a particular problem instance. For any different instance or environment change, one needs to redo the optimisation to get a new solution. On the contrary, hyper-heuristic aims to evolve a heuristic that can perform well on a wide range of problem instances, including unseen (future) instances. First, a general framework of GP as a hyper-heuristic is given. Then, several case studies in job shop scheduling, arc routing problem and travelling thief problem are given to show how to apply the general framework, as well as effective strategies to handle the key challenges and open issues such as feature selection, efficient evaluation, and generalisation.

Speaker's Biography: Dr. Yi Mei is a Lecturer at the School of Engineering and Computer Science, Victoria University of Wellington, Wellington, New Zealand. His research interests include evolutionary computation in scheduling, routing and combinatorial optimisation, as well as evolutionary machine learning, genetic programming, feature selection and dimensional reduction. Yi has more than 40 publications, including the top journals in EC and Operations Research (OR) such as IEEE TEVC, IEEE Transactions on Cybernetics, European Journal of Operational Research, ACM Transactions on Mathematical Software, and top EC conferences (GECCO). As the sole investigator, he won the 2nd prize of the Competition at IEEE WCCI 2014: Optimisation of Problems with Multiple Interdependent Components. He received the 2010 Chinese Academy of Sciences Dean's Award (top 200 postgraduates all over China) and the 2009 IEEE Computational Intelligence Society (CIS) Postgraduate Summer Research Grant (three to four recipients all over the world). Yi is serving as the committee member of IEEE ECTC Task Force on Evolutionary Scheduling and Combinatorial Optimisation, IEEE CIS Task Force on EC for Feature Selection and Construction and IEEE CIS Task Force on Large Scale Global Optimisation. He is a guest editor of the Genetic Programming Evolvable Machine journal, and co-chair of a number of special sessions in international conferences such as IEEE CEC. He is serving as a reviewer of over 20 international journals including the top journals in EC and OR and PC member of almost 20 international conferences.

This presentation will give an introduction on the application of GP in evolving heuristics for combinatorial optimisation problems, which are usually important due to their wide real-world applications.

Speakers in this video

Advertisment

Advertisment