时 间：9月30日16:00
地 点：文理楼580
嘉 宾：Nikhil R. Pal
报告摘要：
“Nature is pleased with simplicity. And nature is no dummy”; “I ask many questions, and when the answer is simple, then God is answering.” . These quotes by the two greatest scientists of all time suggest a very important point: “truth” is simple, the principle of parsimony. Taking inspiration from this, in science and technology when we make a model to explain some natural phenomena we should always look for a simple model that works. In other words, Make everything as simple as possible, but not simpler..
Sparse modelling is a particular manifestation of the principle of parsimony; it is one of the ways to realize “simple models”. The literature on sparse modeling in statistical learning is quite rich, but in fuzzy modelling, although, explicitly or implicitly, sometimes it is used as a good design principle, it is not a very active an area of research. In this talk, first we shall discuss very briefly a few illustrative approaches to classical sparse modeling in the context of clustering and then talk about some of the attempts in fuzzy clustering. This will then be followed by sparse modelling in switching regression problem - Fuzzy Garrotte Switching Regression and Fuzzy Lasso Switching Regression. Finally, we shall discuss how fuzzy rule-based classifier as well as function approximation type systems can be made parsimonious.
嘉宾简介：
Nikhil R. Pal is an INSA Senior Scientist in the Electronics and Communication Sciences Unit at the Indian Statistical Institute (ISI). He is a Fellow of numerous national and international academies and professional societies, including IEEE, TWAS, IFSA, INSA, and the Indian National Academy of Engineering. His research interests include computational intelligence, machine learning, and artificial intelligence. He served as Editor-in-Chief of IEEE Transactions on Fuzzy Systems from 2005 to 2010 and has held editorial and advisory positions for several international journals. He received the 2015 IEEE Computational Intelligence Society Fuzzy Systems Pioneer Award and the 2021 IEEE CIS Meritorious Service Award. He served as IEEE CIS Vice-President for Publications (2013–2016), President (2018–2019), and Distinguished Lecturer during multiple terms.
