In this talk we will discuss the synergy between machine learning and metaheuristics. First we will show how machine learning can help in designing efficient metaheurisitics by guiding the development of the different components of metaheuristics such as the objective function, search operators, initial solutions. The parameter tuning and the problem models are also discussed. In a second time, we will show how metaheuristics can be used to solve machine learning problems. Some examples for clustering, classification, feature selection and association rules are given. Finally we focus on the hyper-parameters optimization of machine learning models.
Prof. El-ghazali Talbi received the Ph.D degrees in Computer Science from the Institut National Polytechnique de Grenoble in France. Since 2001, he is a full Professor at the University of Lille. He is the founder and head of the INRIA Dolphin project. He has many collaborative national, European and international projects. His current research interests are in the field of multi-objective optimization, parallel algorithms, metaheuristics, combinatorial optimization, hybrid and cooperative optimization and learning.
Professor Talbi has to his credit more than 150 publications in journals, books and conferences. He was a guest editor of more than 15 special issues in different journals (Journal of Heuristics, Journal of Parallel and Distributed Computing, European Journal of Operational Research, Theoretical Computer Science, Computers and Operations Research, Journal of Global Optimization). He has an h-index of 52. He is the co-founder and the coordinator of the research group dedicated to Metaheuristics: Theory and Applications (META). He served in different capacities on the programs of more than 100 national and international conferences. His work on metaheuristics (e.g. his book entitled "Metaheuristics: from design to implementation") has a large impact and visibility in the field of optimization.