Public University of Navarra, Spain
In this talk, we will review some recent extensions of the usual notions of Choquet and Sugeno integral. We will see that we recover a wide family of functions, which, although in general do not fulfill the requirements of an aggregation functions, fall into the wider field of pre-aggregation functions.
In particular, we will see how these new functions can be used in some ensemble techniques in order to fuse data from different channels in order to be able to predict from electroencefalographical signals whether a given subject is thinking on the movement olf one hand or another. Furthermore, we will experimentally prove that the use of these extensions of usual fuzzy integrals improve the results obtained by using some other aggregations.
Humberto Bustince Sola is a full professor of Computer Science and Artificial Intelligence at the Public University of Navarra and honorary professor at the University of Nottingham since 2017. He is the main researcher of the Research Group on Artificial Intelligence and Approximate Reasoning, whose research lines are both theoretical (data fusion functions, information and comparison measures, fuzzy sets and their extensions) and applied (Deep learning, image processing, classification, machine learning, data mining, big data or the computational brain). He has led 13 research projects funded by national and regional governments, and two excellence networks on soft computing. He has been the main researcher in projects with companies and entities such as Caja de Ahorros de Navarra, INCITA, Gamesa Tracasa or the Servicio Navarro de Salud. He has taken part in two international projects. He has authored or coauthored more than 300 works, according to Web of Science, including around 160 in Q1 journals. He was a highly cited researcher among the top 1%most relevant scientists in the world in 2018, according to Clarivate Analytics. He collaborates with first line research groups from countries such as United Kingdom, Belgium, Australia, the Czech Republic, Slovakia, Canada or Brasil. He is editor in chief of the Mathware&Soft Computing online magazine of the European Society of Fuzzy Logic and technologies and of the Axioms journal. Associated editor of the IEEE Transactions on Fuzzy Systems journal and member of the editorial boards of the journals Fuzzy Sets and Systems, Information Fusion, International Journal of Computational Intelligence Systems and Journal of Intelligent & Fuzzy Systems. Moreover, he is a coauthor of a book about averaging functions, and has been the co-editor of several books. He has been in charge of organizing several first level international conferences such as EUROFUSE 2009 and AGOP 2013. He is Senior Member of IEEE y Fellow of the International Fuzzy Systems Association (IFSA). Member of the Basque Academy of Sciences, Arts and Literature, Jakiunde, since 2018. He has advised 11 Ph.D thesis.
He was awarded the Cross of Carlos III the Noble by The Government of Navarra in 2017. He got the National Computer Science Prize José García Santesmases in 2019 and the Scientific Excellence Award of EUSFLAT the same year.
IRIT, France
This talk surveys various areas of research initiated by Zadeh when he consistently developed the basic building blocks of fuzzy set theory and approximate reasoning over twenty years. We try to highlight a number of resulting important concepts that emerged from this enterprise. We consider the future of fuzzy sets, indicating promising topics of research and what some, including the author, may view as barren directions. These reflexions also result from experience as co-editor in chief of Fuzzy sets and Systems for two decades.
Didier Dubois is an Emeritus Research Advisor at IRIT, the Computer Science Department of Paul Sabatier University in Toulouse, France and has belonged to the French National Centre for Scientific Resarch (CNRS) since 1984.
He is the co-author, with Henri Prade, of two books on fuzzy sets and possibility theory, and more than 15 edited volumes on uncertain reasoning and fuzzy sets. Also with Henri Prade, he coordinated the HANDBOOK of FUZZY SETS series published by Kluwer (7 volumes, 1998-2000, 2 of which he co-edited). It includes the book Fundamentals of Fuzzy Sets, edited again with H. Prade (Kluwer, Boston, 2000). He co-edited several other books including the edited volume "Decision-making process - Concepts and Methods" (Wiley, 2009) with D. Bouyssou, H. Prade, and M. Pirlot. He has contributed more than 300 technical journal papers on uncertainty theories and applications.
He has been a co-Editor-in -Chief of Fuzzy Sets and Systems for more than 20 years. He is an Advisory Editor of the IEEE Transactions on Fuzzy Systems. He is a member of the Editorial Board of several other technical journals, such as the International Journal on Approximate Reasoning, International Journal of General Systems, and Information Sciences, Journal of Applied Logics, among others.
He is a former president of the International Fuzzy Systems Association (IFSA) (1995-1997) and received the IFSA AWARD (2015) for significant impact on fuzzy logic related research and applications. Besides, in 2002 he received the Pioneer Award of the IEEE Neural Network Society, and, in 2012, the Scientific Excellence Award" from the European Society for Fuzzy Logic and Applications EUSFLAT", and honorary doctorates from the Polytechnic University of Mons, Belgium in 1997, and Obuda University Budapest, Hungary, in 2016.
His topics of interest range from logic-based Artificial Intelligence to Operations Reasearch and Decision Sciences, with emphasis on the modelling, representation and processing of imprecise and uncertain information in reasoning, risk analysis, and problem-solving tasks.
University of Alberta, Canada
Artificial Intelligence (AI) today seems poised to fundamentally reshape modern society. We have reached the point where modern algorithms (e.g. deep learning) can exceed human performance in certain domains. However, they are “black boxes,” and their decision-making process is opaque; sometimes to the point of being incomprehensible to even human experts in the area. It is human nature to distrust what we do not understand, and this leads us to a profound, even existential question for the field of AI: why should humans trust these algorithms with so many important tasks and decisions? Simply put, if human users do not trust AI algorithms, they will not be used.
For over fifty years, AI researchers have argued that giving users a good explanation of why an AI made a decision helps those users to trust the AI more. Research into how those explanations should be created and presented has been going on for all that time; in the last few years, this work has been called “eXplainable Artificial Intelligence,” or XAI. Numerous funding organizations (particularly DARPA in the USA) have also recently made considerable investments in studying XAI. What is perhaps less appreciated, however, is that the concepts behind this topic have been intimately connected to fuzzy logic for the whole history of our field. This talk reviews that long history and demonstrates its connections to the general Artificial Intelligence literature. We then examine how computational intelligence contributes to modern research into explainable systems. Finally, we close with some observations about the current state and possible future directions of research into fuzzy logic and XAI.
Scott Dick received his B.Sc. degree in 1997, his M.Sc. degree in 1999, and his Ph.D. in 2002, all from the University of South Florida. His Ph.D. dissertation received the USF Outstanding Dissertation Prize in 2003. He was an Assistant Professor from 2002 to 2008, an Associate Professor from 2008 to 2018, and has been a Full Professor since 2018, all with the Department of Electrical and Computer Engineering at the University of Alberta in Edmonton, AB. Since 2017, he has also been the Program Director for Computer Engineering.
Dr. Dick’s research interests are in Computational Intelligence, machine learning, data mining, and the application of these technologies to real-world problems (e.g. Smart Grid, livestock disease management, and anti-phishing technologies). He has primarily focused on two topics: firstly, “complex fuzzy logic,” is an extension of type-1 fuzzy logic to complex-valued membership grades. Dr. Dick’s work includes both theoretical analysis of, and constructing machine learning systems based on, this new concept. Secondly, he has worked on constructing eXplainable Artificial Intelligence (XAI) systems using fuzzy logic. His work has been funded by NSERC, SSHRC, the Alberta Science and Research Authority, Alberta Innovates, Hewlett-Packard, PRECARN Inc., and Transport Canada.
Dr. Dick is currently the President (2020-2022) of the North American Fuzzy Information Processing Society. He is a member of the IEEE Computational Intelligence Society’s Fuzzy Systems Technical Committee, and was the chair of the FSTC Task Force on Complex Fuzzy Sets & Logic from 2014-2018. He is an Associate Editor for Evolving Systemsand Complex and Intelligent Systems. He is a member of the ACM, IEEE, and ASEE.