Prof. O. Cordon
Transkrypt
Prof. O. Cordon
Wykłady Prof. Oscara Cordóna z Hiszpanii Genetic Fuzzy Systems. Fuzzy Knowledge Extraction by Evolutionary Algorithms 9-10 marca 2009 r. Prof. Oscar Cordón pracuje w European Centre for Soft Computing w Mieres. Jest jednym z najlepszych specjalistów w zakresie inteligencji obliczeniowej, a w szczególności genetycznych systemów rozmytych. Jest autorem 210 publikacji, które uzyskały ponad 900 cytowań w SCI. Program pobytu Prof. Oscara Cordóna na Politechnice Wrocławskiej: Poniedziałek 9.03.2009, godz. 9.15-11.00, sala 4.48 – B4 Seminarium dla pracowników naukowych: Genetic Algorithms and Soft Computing in Forensic Identification (Algorytmy genetyczne i miękkie obliczenia w antropologii sądowej) Abstract: In this talk, we will describe the most recent developments of a coordinated project between the European Centre for Soft Computing and the University of Granada’s Physical Anthropology Lab where soft computing techniques are being applied to forensic identification. In particular, the procedure followed by a forensic anthropologist to identify missing people through photographic supra-projection is being automated. We will show how fuzzy logic and evolutionary algorithms can be used for the automatic reconstruction of tridimensional models of human skulls and for the automatic superimposition of those skull 3D models with pictures of the missing people. The results obtained in several real-world cases will be reported. Krótka biografia: Oscar Cordón received his M.S. degree (1994) and his Ph.D. (1997) both in Computer Science from the University of Granada, Spain, where he has been professor at the Dept. of Computer Science and Artificial Intelligence. Since April 2006, he is the Principal Researcher of the "Applications of Fuzzy Logic and Evolutionary Algorithms" research unit at the European Centre for Soft Computing, located in Mieres, Asturias. Last December, 2008, he received the Full Professor accreditation from the Spain’s National Quality Evaluation Agency. Prof. Cordón has a wide international experience in the area of soft computing, specifically in the fundamentals and applications of evolutionary algorithms and other metaheuristics, as well as in fuzzy systems. He has published more than 210 peer-reviewed scientific publications, including a co-authored book on genetic fuzzy systems and 47 SCI-JCR-indexed journal papers. In February, 5, 2009, his publications had received 917 citations, carrying an h-index of 17, and being included in the 1% of most cited researchers in the world (source: Thomson’s Web of Knowledge; h-index=27 in Google Scholar). He has participated in 24 national and European research projects and contracts, coordinating 9 of them. Dr. Cordón is associate editor and editorial board member of six scientific journals. He also created the Genetic Fuzzy Systems Task Force, Fuzzy Systems Technical Committee, IEEE Computational Intelligence Society, in 2004 and chaired it till 2007. Moreover, he is a member of the IPC of more than 80 conferences and frequently acts as a reviewer for more than 20 international journals. His current main research interests are in the fields of soft computing for forensic anthropology and medical imaging; genetic fuzzy systems; soft computing and visual science maps; and evolutionary computation, ant colony optimization and other metaheuristics. Poniedziałek 9.03.2009, godz. 11.15-13.00, sala 409 – B4 Otwarty wykład dla studentów i doktorantów. Lecture 1’s Title: Brief Introduction to Fuzzy Rule-Based Systems and Genetic Algorithms Abstract: This first part of the lecture will be devoted to introduce the basics of fuzzy rule-based systems and genetic algorithms, the two constituents of genetic fuzzy systems. First, the topic of system identification with fuzzy logic techniques will be reviewed, and the different kinds of fuzzy models and classifiers will be described, mainly focusing on the fuzzy rule structure and the inference mechanism. Wang and Mendel’s heuristic method to derive the fuzzy rule base will be also presented. Then, the main principles of genetic algorithms will be shown, starting from their natural inspiration, reviewing their components, analyzing some design issues (such as coding, genetic operators, evolution model, …) and providing some applications. Poniedziałek: 9.03.2009, godz. 13.15-15.00, sala 4.48 – B4 Otwarty wykład dla studentów i doktorantów. Lecture 2’s Title: Genetic Fuzzy Systems: Brief Introduction and Historical Development. Classical models for the Evolutionary Tuning of Fuzzy Rule-based Systems Abstract: In the second part, the topic of genetic fuzzy systems, composed of soft computing techniques augmenting the approximate reasoning method of fuzzy systems with the learning abilities of evolutionary algorithms, will be introduced. A simple taxonomy of these approaches based on the pioneering genetic fuzzy rule-based systems proposals that comprised the birth of the area in 1991 will be described. Then, the historical evolution it suffered will be reviewed. Finally, the basics of evolutionary tuning processes will be analyzed. Wtorek 10.03.2009, godz. 9.15-11.00, sala 4.48 – B4 Otwarty wykład dla studentów i doktorantów: Lecture 3’s Title: Classical models for the Evolutionary Learning of Fuzzy Rule-based Systems. Some Real-World Applications Abstract: The third part deals with the main branch of study of genetic fuzzy systems, fuzzy knowledge base learning by means of evolutionary algorithms. The three classical genetic learning models, Michigan, Pittsburgh, and Iterative Rule Learning, will be first introduced, as well as different genetic fuzzy systems taking them as a base to learn the fuzzy rule base. The very important topic of genetic selection of fuzzy rules to decrease the fuzzy system complexity will be studied then. Next, the problematic associated to the evolutionary learning of the whole knowledge base will be posed. Finally, some real-world applications solved with genetic fuzzy systems will be reviewed. Wtorek 10.03.2009, godz. 11.15-13.00, sala 4.48 – B4 Otwarty wykład dla studentów i doktorantów: Lecture 4’s Title: Advanced Genetic Fuzzy Systems approaches. What’s Next? Abstract: The aim of the fourth and last part of the lecture is to introduce the attendees to some of the existing advanced evolutionary fuzzy knowledge base derivation techniques, as well as the modern genetic tuning approaches, developed in the last few years. Hot topics in the area such as the fuzzy system interpretability-accuracy trade-off will be studied and some future lines of development of the discipline will be identified.