Explainable AI: Data-driven Fuzzy Systems
Distinguished Professor, National Chung-Hsing University, Taiwan
AI has become a popular research topic in recent years and has shown great success in different applications. However, most AI models function as black boxes and it is hard to explain the inference process of a suggestion made by these models. In this context, explainable AI (XAI) has attracted the attention of many researchers. Fuzzy systems (FSs) that show the advantage of interpretability in their inference fuzzy rules may provide a possible solution to XAI. In this talk, I will introduce our recent research results in data-driven interpretable FSs. Two learning techniques of data-driven interpretable FSs, including fuzzy neural networks (FNNs) and multiobjective evolutionary FSs (EFSs), will be introduced together with their applications. For FNNs, I will start with learning with low-scale data and its application to estimate the severity of obstructive sleep apnea (OSA). Learning of FNNs with high-scale feature maps from a deep learning model and its application to image classification problems will then be given. The technique of multiobjective EFSs aims to find a set of non-dominated FSs that show tradeoffs between different objectives such as system interpretability and model accuracy through multiobjective evolutionary computation algorithms. In this subtopic, I will introduce the Multiobjective EFS we recently proposed and its application to evolutionary mobile robot control. To boost the learning efficiency of multiobjective EFSs, the technique of reinforcement neural fuzzy surrogate-assisted learning will be given at the end of this talk.
Chia-Feng Juang received the B.S. and Ph.D. degrees in Control Engineering from the National Chiao-Tung University, Hsinchu, Taiwan, R.O.C., in 1993 and 1997, respectively. Since 2001, he has been with the Department of Electrical Engineering, National Chung-Hsing University (NCHU), Taichung, Taiwan, R.O.C., where he became a Full Professor in 2007 and has been a Distinguished Professor since 2009. He served as the Chapter Chair of IEEE Computational Intelligence, Taipei Chapter, in 2017-2018, during which the chapter won the Outstanding Chapter Award from IEEE Taipei Session. Dr. Juang has authored or coauthored nine book chapters, 100 journal papers (including over 55 IEEE journal papers), and over 120 conference papers. His current research interests include computational intelligence, intelligent control, computer vision, and evolutionary robots. Five of his highly-cited papers have collected 1900+ (3000+) citations in Web of Science (Google Scholar).
Dr. Juang received the Outstanding Youth Award from Taiwan System Science and Engineering Society, Taiwan, in 2010; the Excellent Research Award from NCHU, Taiwan, in 2010; the Outstanding Youth Award from Taiwan Fuzzy Systems Association, Taiwan, in 2014; and the Outstanding Automatic Control Engineering Award from Chinese Automatic Control Society (CACS), Taiwan, in 2014. He was elevated to CACS Fellow in 2016 and IEEE Fellow in 2019. He is an Associate Editor of the IEEE TRANSACTIONS ON FUZZY SYSTEMS, the IEEE TRANSACTIONS ON CYBERNETICS, the Asian Journal of Control, and the Journal of Information Science and Engineering and an Area Editor of the International Journal of Fuzzy Systems.