ICAISC 2010 Invited Talks
|
|
|
|
Invited Talk
"Complex-Valued Neurons With Phase-Dependent Activation Functions"
Igor Aizenberg (Biography)
Texas A&M University-Texarkana
USA
|
|
|
|
|
|
Invited Talk
"A new implemenation of the coVAT algorithm for visual assessment of clusters in
rectangular relational data"
Jim Bezdek (Personal web page)
University of West Florida
USA
|
|
|
|
|
|
Invited Talk
"Advances in Meta Learning"
Włodzisław Duch (Biography)
Nicolaus Copernicus University
Poland
|
|
|
|
|
|
Invited Talk
"Affective Brain Machine Interfaces and Multi-way Analysis of Human Brain Data"
Andrzej CICHOCKI (Personal web page)
Riken Brain Science Institute
JAPAN
Summary
One challenge in computational and system neuroscience is to understand the relationship between perception, cognition, emotions and social interactions. The challenge is to recognize and discriminate human emotions and design an intelligent computational system for rehabilitation purpose. Research is this direction is motivated by the fact that emotions pervades human life and often motivates human behavior. Emotional cues play an important role in forecasting human mental states, human attention, performance (e.g., hazard in driving a car) and future actions. We attempt to develop and design a system which could interpret and recognize human emotions and related affective phenomena via experiments using affective (emotional) BMI (i.e., control directly by brain signals, for example, a wheelchair, a robot arm, interactive games, etc.). Brain machine interface is a new and promising paradigm in neuroscience. BMI allows us to investigate in real time correlations between brain activates, perceptions, mental tasks, cognition and actions. Moreover, we can learn from BMI how information from different sensory streams is processed and integrated in the brain and use this knowledge to build efficient bio-engineering devices. We are performing some basic experiments, by simultaneously recording EEG/fNIRS signals for two or even 4 subjects during e.g., watching video-clips, hearing music and social interactions.
In order to study of neural basis of perception, expression and regulation of emotions and social interactions, we need new technologies and machine learning tools. Actually, we attempt to extract 'unseen' hidden components, common factors and/or regularities by applying multiway analysis and tensor factorization models. In this talk, we will review different models of multi way blind sources separation and tensor decompositions, including NMF/NTF (Nonnegative Matrix/Tensor Factorization), multiway ICA (Independent Component Analysis), SCA (Sparse Component Analysis), and Morphological Component Analysis (MCA), with special emphasis of efficient regularization and model selection issues. We focus on finding common patterns, and hidden components for a very large scale multidimensional data (tensors). We will address the important issue how to determine or extract hidden variables and how to detect normal and abnormal behavior in EEG/ECoG brain data. We will discuss how to extract spatio-spectral and temporal information from the human brain and how to perform categorizations, classifications and multiway clustering.
Brief Biography
Andrzej
Cichocki received the M.Sc. (with honors), Ph.D. and Dr.Sc. (Habilitation)
degrees, all in electrical engineering. from Warsaw University of
Technology (Poland). Since 1972, he has been with the Institute of Theory of Electrical
Engineering, Measurement and Information Systems, Faculty of Electrical Engineering at
the Warsaw University of Technology, where he obtain a title of a full
Professor in 1995. He spent several years at University Erlangen-Nuerenberg (Germany), at the Chair of Applied and Theoretical Electrical Engineering, as an
Alexander-von-Humboldt Research Fellow and Guest Professor. In 1995-1997 he
was a team leader of the laboratory for Artificial Brain Systems, at
Frontier Research Program RIKEN (Japan), in the Brain Information
Processing Group. He is currently team leader of the laboratory for Advanced Brain Signal Processing,
at RIKEN Brain Science Institute (JAPAN). He is
co-author of more than 250 technical papers and 4 monographs (two of them
translated to Chinese): Nonnegative
Matrix and Tensor Factorizations: Applications to Exploratory
Multi-way Data Analysis, John Wiley-2009; Adaptive Blind Signal
and Image Processing (co-authored with Professor
Shun-ichi Amari ;Wiley, April 2003 -revised edition), CMOS
Switched-Capacitor and Continuous-Time Integrated Circuits and Systems
(co-authored with Professor Unbehauen; Springer-Verlag, 1989) and Neural
Networks for Optimizations and Signal Processing
(Teubner-Wiley,1994). He is Editor in Chief of International Journal
Computational Intelligence and Neuroscience.
|
|
|
|
|
|
Invited Talk
"A General Framework for Representation and Training of Dynamic Neural Networks"
Martin Hagan (Biography)
School of Electrical and Computer Engineering
Oklahoma State University
USA
|
|
|
|
|
|
Invited Talk
"Recent progress in applications of complex-valued neural networks"
Akira Hirose (Biography)
University of Tokyo
Japan
|
|
|
|
|
|
Invited Talk
"Greed and fairness in choice and decision making: a neuroeconomic perspective"
Janusz Kacprzyk (Biography)
Polish Academy of Sciences
Poland
|
|
|
|
|
|
Invited Talk
"Recognition Technology in Eldercare"
Jim Keller (Biography)
University of Missouri-Columbia
USA
|
|
|
|
|
|
Invited Talk
"Neuronal systems biology with functional brain imaging"
Jagath C. Rajapakse (Biography)
BioInformatics Research Center
Nanyang Technological University
Singapore
|
|
|
|
|
|
Invited Talk
"Biometric Technologies of Russian Biometric Society and Their Implementations"
Igor Spiridonov (Biography)/ Andrey Khrulev
Bauman University, Moscow
Russia
|
|
|
|
|
|
Invited Talk
"The State of the Art of Neurodynamic Optimization"
Jun Wang (Biography)
Department of Mechanical and Automation Engineering
The Chinese University of Hong Kong
Hong Kong
|
|