Dr. Yaochu Jin
Honda Research Institute Europe, Germany
Machine learning is inherently a multi-objective task. Traditionally, however, either only one of the objectives is adopted as the cost function or multiple objectives are aggregated to a scalar cost function. This can mainly attributed to the fact that most conventional learning algorithms can only deal with a scalar cost function. Over the last decade, efforts on solving machine learning problems using the Pareto-based multi-objective optimization methodology have gained increasing impetus, particularly thanks to the great success of multi-objective optimization using evolutionary algorithms and other population-based stochastic search methods. It has been shown that Pareto-based multi-objective learning approaches are more powerful compared to learning algorithms with a scalar cost functions in addressing various topics of machine learning, such as clustering, feature selection, improvement of generalization ability, knowledge extraction, and ensemble generation.
This talk provides first a brief overview of Pareto-based multi-objective machine learning techniques. In addition, a number of case studies are provided to illustrate the major benefits of the Pareto-based approach to machine learning, e.g., how to identify interpretable models and models that can generalize on unseen data from the obtained Pareto-optimal solutions. Three approaches to Pareto-based multi-objective ensemble generation are compared and discussed in detail. Most recent results on multi-objective optimization of spiking neural networks will be presented.
Yaochu Jin received the B.Sc., M.Sc., and Ph.D. degrees from Zhejiang University, Hangzhou, China, in 1988, 1991, and 1996, respectively, and the Ph.D. degree from Ruhr-Universität Bochum, Bochum, Germany, in 2001.
Dr. Jin is currently a Principal Scientist with the Honda Research Institute Europe, Offenbach, Germany. He was an Associate Professor of the Electrical Engineering Department, Zhejiang University, Hangzhou, China, a Visiting Researcher and a Scientific Staff of the Institut für Neuroinformatik, Ruhr-Universität Bochum, Germany, a Postdoctoral Associate of the Industrial Engineering Department, Rutgers, Piscataway, USA, and a Scientist with the Honda R&D Europe, Offenbach, Germany. His research interests include computational approaches to understanding evolution and learning in biology, and evolutionary and learning approaches to complex systems design. He is the editor of Evolutionary Computation in Dynamic and Uncertain Environments (Berlin, Springer, Germany, 2007), Multi-Objective Machine Learning (Berlin, Germany: Springer, 2006) and Knowledge Incorporation in Evolutionary Computation (Berlin, Germany: Springer, 2005), the author of the book Advanced Fuzzy Systems Design and Applications (Heidelberg, Germany: Springer, 2003), and the author of over 80 journal and conference papers.
Dr. Jin is currently an Associate Editor of the IEEE Transactions on Neural Networks, the IEEE Transactions on Control Systems Technology, the IEEE Transactions on Systems, Man, and Cybernetics, Part C, and the IEEE Computational Intelligence Magazine. He is also an Editorial Member of the Soft Computing Journal and the International Journal of Cognitive Informatics and Natural Intelligence. He has been a Guest Editor of five journal special issues, including one on Evolutionary Optimization in the Presence of Uncertainties in the IEEE Transactions on Evolutionary Computation. He is the Program Co-Chair of the 2004 Int. Conference on Fuzzy Systems and Knowledge Discovery, Changsha, China, and 2007 IEEE Symposium on Multi-Criteria Decision Making, Honolulu, Hawaii. Dr. Jin is a Senior Member of IEEE.