Keynote Speakers

Professor Nikola Kasabov

Evolving Intelligent Systems: Methods and Applications

Abstract

This presentation gives some background information and novel approaches to building evolving intelligent systems (EIS), along with their applications in speech and image processing, robotics, bioinformatics and brain study. The EIS evolve their structure and functionality through learning from data in both on-line and off-line incremental modes, supervised and unsupervised modes, and facilitate data and knowledge integration, rule extraction and rule manipulation. Evolving Connectionist Systems (ECOS) implement the main ideas and principles of EIS. The evolving process in ECOS is defined by parameters, "genes" [1]. ECOS are characterized by local, clustering-based learning. ECOS extend further the classical knowledge-based neural networks [2]. Evolutionary computation methods are also applied to optimize ECOS parameters over time. ECOS are implemented as part of a general- purpose data mining and knowledge discovery environment NeuCom that will be used for demonstrations (see www.theneucom.com). New methods for EIS are presented, such as: transductive, personalised modelling [3]; incremental feature selection [4]; ensemble learning [5].

ECOS, as well as other techniques for EIS, are demonstrated in the talk on challenging knowledge engineering problems, such: adaptive, on-line speech and image processing from streams of data [6]; adaptive mobile robots; adaptive time-series prediction [3]. ECOS are also demonstrated on challenging problems in Bioinformatics and Neuroinformatics, such as: dynamic modeling of gene regulatory networks (GRN) [7]; adaptive medical prognostic systems [8]; adaptive models of human perception [1]; computational neurogenetic modelling using spiking neural networks [9]. Further directions of EIS include: new methods and theories for evolving intelligence; dynamic evolving neurogenetic models; modeling complex brain processes and diseases; whole cell modeling; hardware implementation coupled with multi-sensor information processing.

Keywords: Evolving intelligent systems, Evolving connectionist systems, Adaptive robotics, Adaptive speech and image processing, Neuroinformatics, Bionformatics, Knowledge-based neural networks, Gene regulatory networks, Computational neurogenetic modeling.

References
[1] N.Kasabov, Evolving Connectionist Systems: Methods and Applications in Bioinformatics, Brain Study and Intelligent Machines, Springer Verlag, 2002 (www.springer.de)
[2] N.Kasabov, Foundations of neural networks, fuzzy systems and knowledge engineering, MIT Press, 1996 (www.mitpress.edu)
[3] Q. Song and N. Kasabov, TNFI: A Neuro-Fuzzy Inference Method for Transductive Reasoning, IEEE Transactions on Fuzzy Systems, December, vol.13, issue 6, 2005, 799-808.
[4 ] S. Pang, S. Ozawa and N. Kasabov, Incremental Linear Discriminnant Analysis for Classification of Data Streams, IEEE Trans. SMC-B, vol. 35, No. 5, 2005, 905 - 914
[5] Chan S.H., Kasabov N., Fast Neural Network Ensemble Learning via Negative-Correlation Data Correction, IEEE Transactions on Neural Networks, December, 2005
[6] S. Ozawa, S.Too, S.Abe, S. Pang and N. Kasabov, Incremental Learning of Feature Space and Classifier for Online Face Recognition, Neural Networks, August, 2005
[7] N Kasabov, I.A. Sidorov, D S Dimitrov, Computational Intelligence, Bioinformatics and Computational Biology: A Brief Overview of Methods, Problems and Perspectives, Journal of Computational and Theoretical Nanoscience, Vol. 2 No. 4, December, 2005, 473-491.
[8] Q. Song, N. Kasabov, T. Ma, M. Marshall, Integrating regression formulas and kernel functions into locally adaptive knowledge-based neural networks: a case study on renal function evaluation, Artificial Intelligence in Medicine, February, 2006
[9] N. Kasabov, L. Benuskova L and Wysoski SG (2005) Computational neurogenetic modeling: integration of spiking neural networks, gene networks, and signal processing techniques. In: ICANN 2005, LNCS 3697, W. Duch et al (Eds), Springer-Verlag, Berlin Heidelberg, pp. 509-514.

Nikola Kasabov

Biography

Professor Nikola Kasabov is the Founding Director and the Chief Scientist of the Knowledge Engineering and Discovery Research Institute KEDRI, Auckland (www.kedri.info/). He holds a Chair of Knowledge Engineering at the School of Computer and Information Sciences at Auckland University of Technology. He is a Fellow of the Royal Society of New Zealand, Fellow of the New Zealand Computer Society and a Senior Member of IEEE. He is a member of the Board of Governors of the International Neural Network Society (INNS), of the Asia Pacific Neural Network Assembly (APNNA), and on several technical committees of the IEEE Computational Intelligence Society. Kasabov is on the editorial boards of several international journals, that include IEEE Tr NN, IEEE Tr II, Information Science. He chaired the series of ANNES conferences (1993-2001) and is the chair of the NCEI conference series (2002 -). Kasabov holds MSc and PhD from the Technical University of Sofia. His main research interests are in the areas of intelligent information systems, soft computing, neuro-computing, bioinformatics, brain study, speech and image processing, novel methods for data mining and knowledge discovery. He has published more than 320 publications that include 15 books, 90 journal papers, 50 book chapters, 25 patents and numerous conference papers. He has extensive academic experience at various academic and research organisations: University of Otago, New Zealand; University of Essex, UK; University of Trento, Italy; Technical University of Sofia, Bulgaria; University of California at Berkeley; RIKEN Brain Science Institute, Tokyo; University of Kaiserslautern. More information of Prof. Kasabov can be found on the Web site: http://www.kedri.info.