Plenary Lecture
Interval Analysis Strategies and Applications in Multilayer Neural Computing
Professor Dimitrios A. Karras
Sterea Hellas Institute of Technology
Dept. Automation
Greece
E-mail: dakarras2010@gmail.com
Abstract: This speech presents strategies and new tools of interval analysis and their possible applications in neural computation. More specifically, this speech focuses on defining aspects of Multilayer Neural Network training procedures that could be efficiently managed through interval analysis methodologies. Special attention is payed to determining good initial conditions for an algorithm used to train a neural network. This problem is considered as a parameter estimation problem dealing with uncertainty about the initial weights. Interval Analysis approaches model uncertainty in parameter estimation problems using intervals and formulating tolerance problems. Solving a tolerance problem is defining lower and upper bounds of the intervals so that the system functionality is guaranteed within predefined limits. One of the aims of this speech is to show and explain how the problem of determining the initial weight intervals of a neural network can be defined in terms of solving a linear interval analysis problem. The proposed Linear Interval Tolerance Approach copes with uncertainty about the initial weights without any apriori knowledge on input data distribution as required by approaches such fuzzy sets or rough sets. The proposed method is tested on a number of well known benchmarks for neural networks trained with the back-propagation family of algorithms. Its e fficiency is evaluated with regards to standard performance measures and the results obtained are compared against results of a number of well known and established initialization methods. Guidelines and directions are given on how to extend these results, also, to other aspects of neural computation training problems too.
Brief Biography of the Speaker: Dimitrios A. Karras received his Diploma and M.Sc. Degree in Electrical and Electronic Engineering from the National Technical University of Athens, Greece in 1985 and the Ph. Degree in Electrical Engineering, from the National Technical University of Athens, Greece in 1995, with honours. From 1990 and up to 2004 he collaborated as visiting professor and researcher with several universities and research institutes in Greece. Since 2004, after his election, he has been with the Sterea Hellas Institute of Technology, Automation Dept., Greece as associate professor in Digital Systems and Signal Processing as well as with the Hellenic Open University, Dept. Informatics as a visiting professor in Communication Systems (the latter since 2002 and up to 2010). He has published more than 65 research refereed journal papers in various areas of pattern recognition, image/signal processing and neural networks as well as in bioinformatics and more than 180 research papers in International refereed scientific Conferences. His research interests span the fields of pattern recognition and neural networks, image and signal processing, image and signal systems, biomedical systems, communications, networking and security. He has served as program committee member in many international conferences, as well as program chair and general chair in several international workshops and conferences in the fields of signal, image, communication and automation systems. He is, also, former editor in chief (2008-2015) of the International Journal in Signal and Imaging Systems Engineering (IJSISE), academic editor in the TWSJ, ISRN Communications and the Applied Mathematics Hindawi journals as well as associate editor in various scientific journals. He has been cited in more than 1500 research papers, his H/G-indices are 16/29 (Google Scholar) and his Erdos number is 5. His RG score is 30.33 (https://www.researchgate.net/profile/Dimitrios_Karras2/)