mlsplogo MLSP2013
IEEE International Workshop on
Machine Learning for Signal Processing

September 22-25, 2013  Southampton, United Kingdom

Nonlinear System Identification in Reproducing Kernel Hilbert Spaces: Methods and Applications
Cédric Richard Cédric Richard
Laboratoire Lagrange, University of Nice Sophia-Antipolis, France
Personal homepage


Cédric Richard received the Dipl.-Ing. and the M.S. degrees in 1994 and the Ph.D. degree in 1998 from the University of Technology of Compiègne, France, all in electrical and computer engineering. From 1999 to 2003, he was an Associate Professor at the University of Technology of Troyes (UTT), France. From 2003 to 2009, he was a Professor at UTT. Since september 2009, he has been a Professor in the Lagrange Laboratory, University of Nice Sophia-Antipolis, France. In winter 2009, and autumns 2010 and 2011, he was a Visiting Researcher with the Department of Electrical Engineering, Federal University of Santa Catarina (UFSC), Florianopolis, Brazil, to collaborate with Prof. Jose-Carlos M. Bermudez. He is a junior member of the Institut Universitaire de France since October 2010.

His current research interests include statistical signal processing and machine learning.

He was the General Chair of the 21st Francophone conference GRETSI on Signal and Image Processing in Troyes, France, in 2007, and of the IEEE Statistical Signal Processing Workshop (IEEE SSP'11) in Nice, France, in 2011. Since 2005, He has been a member of GRETSI association board of the EURASIP society. He served as an associate editor of the IEEE Transactions on Signal Processing since 2009, and as an Associate Editor of Elsevier Journal in Signal Processing Elsevier. He has been an EURASIP liaison local officer, and a member of the Signal Processing Theory and Methods (SPTM) Technical Committee of the IEEE Signal Processing Society.

Together with Paul Honeine and he won the Best Paper Award for "Solving the preimage problem in kernel machines: a direct method" at the 2009 IEEE International Workshop on Machine Learning for Signal Processing.


Dynamic system modeling has played a crucial role in the development of techniques for stationary and non-stationary signal processing. Most existing approaches focus on linear models due to their inherent simplicity from conceptual and implementational points of view. However, there are many practical situations where the nonlinear processing of signals is needed. Since the pioneering works of Aronszajin, function approximation methods based on reproducing kernel Hilbert spaces (RKHS) have gained wide popularity. In this talk, we will review some recent developments in online techniques for nonlinear system identification in RKHS. We will also focus on their stochastic behavior analysis. Finally, applications in biomedical engineering, sensor networks and hyperspectral data unmixing will be presented.  

Tensor Tools for Signal Processing and Data Analysis
Lieven De Lathauwer Lieven De Lathauwer
ESAT/SCD, Katholieke Universiteit Leuven, Belgium
Personal homepage
Download lecture slides


He obtained the Master's degree in Electro-Mechanical Engineering and the Ph.D. degree in Applied Sciences from KU Leuven, in 1992 and 1997, respectively. From 2000 to 2007 he was Research Associate with the French Centre National de la Recherche Scientifique. Since 2007 he is Associate Professor with KU Leuven. He is affiliated with the Group Science, Engineering and Technology of Kulak, with the group SCD-SISTA of the Electrical Engineering Department (ESAT) and with iMinds Future Health Department. He is Associate Editor of the SIAM Journal on Matrix Analysis and Applications and has served as Associate Editor for the IEEE Transactions on Signal Processing. He was chair of the Workshop on Tensor Decompositions and Applications (TDA 2005), Luminy, France, and of the Workshop on Tensor Decompositions and Applications (TDA 2010), Bari, Italy.

L. De Lathauwer's research concerns the development of tensor tools for engineering applications. It centers on the following axes: algebraic foundations, numerical algorithms, generic methods for signal processing and data analysis, and particular applications.


For more than 20 years, decompositions of higher-order tensors have played a key role in research on independent component analysis and blind source separation. Nowadays tensors are intensively studied in many disciplines. They open up remarkable new possibilities in signal processing, data mining, machine learning, system modelling, scientific computing, statistics, array processing, wireless communication, audio and image processing, biomedical applications, bio-informatics, and so on. On the other hand, tensor methods have firm roots in multilinear algebra, algebraic geometry, numerical mathematics and optimization.

This tutorial is meant to give a sound introduction to the subject. We also discuss new trends and perspectives. We illustrate some of the ideas using Tensorlab, a Matlab toolbox for tensors and tensor computations that we have recently released.  

Powered by CONWIZ, © Copyright 2013. Page maintained by Jan Larsen (Email)