Speaker: Academician Leszek Rutkowski
Title: Probabilistic neural networks for incremental learning over time - varying streaming data
Time: 14:30pm, May 25, 2023 (Thursday)
Website: Teams Link
Abstract:
This talk presents a novel algorithm for incremental learning over streaming data in a non-stationary environment. The idea refers to the applicability of Probabilistic Neural Networks (PNNs), commonly used as a fast and robust method for solving classification problems in a stationary scenario. It is well known that PNNs have solid mathematical foundations but they fail in the non-stationary scenario and are not able to deal with concept drift. In this presentation, the method based on the orthogonal series expansions of unknown probability densities is outlined. This approach significantly extends the applicability of the PNNs. The nonparametric procedures called Incremental Probabilistic Neural Networks (IPNNs) for tracking drifting probability densities or conditional class densities, in the case of classification, are discussed in detail. Their convergence is shown as the stream data size is growing to infinity. Compared to the existing literature, almost exclusively based on various heuristics, the proposed method is mathematically justified and has solid theoretical foundations.
Personal Introduction:
Leszek Rutkowski received the M.Sc. , Ph.D. and D.Sc. degrees from the Wrocław University of Technology, Wrocław, Poland, in 1977, 1980 and 1986, respectively, and the Honoris Causa degree from the AGH University of Science and Technology, Kraków, Poland, in 2014. He is a Honorary Professor of Czestochowa University of Technology, Poland, and serves as a Full Professor in the Institute of Information Technology at the University of Social Sciences, Łódź, Poland. He is also cooperating with the Systems Research Institute of the Polish Academy of Sciences, Warsaw, Poland, and with the Institute of Computer Science, AGH University of Science and Technology, Krakow, Poland, in the area of artificial intelligence and soft computing. From 1987 to 1990, he held a visiting position with the School of Electrical and Computer Engineering, Oklahoma State University, Stillwater, OK, USA. His research interests include data stream mining, big data analysis, neural networks, agent systems, fuzzy systems, image processing, pattern classification, and expert systems. He has published more than 300 technical papers, including more than 30 in various series of IEEE Transactions. He is the author of the following books: Computational Intelligence (Springer, 2008, available also in Polish and Russian), New Soft Computing Techniques for System Modeling, Pattern Classification and Image Processing (Springer, 2004), Flexible Neuro-Fuzzy Systems (Kluwer Academic, 2004), Methods and Techniques of Artificial Intelligence (2005, in Polish), Adaptive Filters and Adaptive Signal Processing (1994, in Polish), and co-author of two others in Polish: Neural Networks, Genetic Algorithms and Fuzzy Systems (1997) and Neural Networks for Image Compression (2000). His recent monograph on stream data mining has been published by Springer in the series Studies in Big Data. He is the president and founder of the Polish Neural Networks Society. He organized and served as a General Chair of the International Conferences on Artificial Intelligence and Soft Computing held in 1996, 1997, 1999, 2000, 2002, 2004, 2006, 2008, 2010, and annulally in 2012 -2022. He was an associate editor of the IEEE Transactions on Neural Networks (1998-2005) and IEEE Systems Journal (2007-2010), Evolving Systems (2017-2019) and few others. He is an editor-in-chief of the Journal of Artificial Intelligence and Soft Computing Research, and he is on the editorial board of the IEEE Transactions on Cybernetics, International Journal of Neural Systems, International Journal of Applied Mathematics and Computer Science , International Journal of Biometric, Knowledge and Information Systems. He is a recipient of the IEEE Transactions on Neural Networks 2005 Outstanding Paper Award. He served in the IEEE Computational Intelligence Society as the chair of the Distinguished Lecturer Program (2008-2009) and the chair of the Standards Committee (2006-2007). He is the founding chair of the Polish chapter of the IEEE Computational Intelligence Society, which won the 2008 Outstanding Chapter Award. In 2004, he was awarded by the IEEE Fellow membership grade for contributions to neurocomputing and flexible fuzzy systems. He received a degree honoris causa from the prestigious AGH University of Science and Technology in Cracow “in recognition of outstanding scientific achievements in the field of artificial intelligence - in particular, neuro-fuzzy systems”. He is a Full Member of the Polish Academy of Sciences, elected in 2016, and Member of Academia of Europea, elected in 2022.
[Editor: Xiaohan Liu]