PAPERS
Call for Papers
Special Session on Bioinformatics
Special Session on Aeronautics and Space
Submission Schedule
Submission of camera-ready accepted papers
Presentation Guideline

CALL FOR PAPERS

Download call for paper (PDF).

Thanks to the sponsorship of IEEE Signal Processing Society and IEEE Neural Network Council, the thirteenth of a series of IEEE workshops on Neural Networks for Signal Processing will be held in Toulouse, France.

The workshop will feature keynote addresses, technical presentations and panel discussions. Papers are solicited for, but not limited to, the following areas:

Algorithm and Architectures:
Artificial neural networks, kernel methods, committee models, Gaussian processes, independent component analysis, advanced (adaptive, nonlinear) signal processing, (hidden) Markov models, Bayesian modeling, parameter estimation, generalization, optimization, design algorithms.
Applications:
Speech processing, image processing (computer vision, OCR), multimodal interactions, multi-channel processing, intelligent multimedia and web processing, robotics, sonar and radar, bio-medical engineering, bioinformatics, financial analysis, time series prediction, blind source separation, data fusion, data mining, adaptive filtering, communications, sensors, system identification, and other signal processing and pattern recognition applications.
Implementations:
Parallel and distributed implementation, hardware design, and other general implementation technologies.

 

SPECIAL session on bioinformatics

Recent advancements in molecular biology have brought us the possibility of sequencing the genomes of a number of different organisms. These genomes are made up of a large amount of data, billions of characters to be deciphered and processed. New techniques have been developed to process the genomic and proteomic information, from the sequencing of the DNA and the identification of polymorphisms to the search of protein functionality and interaction. The field that aims to study, analyze, process and find appropriate ways of storing and retrieving genomic and proteomic data, is called Bioinformatics.

The special session is open to all persons working at the frontier between Bioinformatics and Signal Processing/Neural Networks and offers them the possibility of presenting and discussing recent and original work that merges the two fields with application to the following areas of research:

For further information contact the organizer:
Dr. Edgardo Ferran
Head of Bioinformatics,
Molecular and Functional Genomics Department,
Sanofi-Synthelabo Recherche, Labege Innopole - BP 137,
31676 Labege Cedex, France.
e-mail: Edgardo.Ferran@sanofi-synthelabo.com

 

SPECIAL session on Aeronautics and Space

Neural networks provide a general framework to solve problems of pattern recognition, non linear regression and optimization, and provide attractive solutions to problems in aeronautics and in space research and development. In these fields, classical methods such as linear methods, are commonly employed with a priori estimation of modeling errors. When experimental data are available, neural learning provides a more flexible and precise identification tool. Another attractive feature of the neural architecture is its modularity. In aeronautics and space, design is performed using complex simulators that consist of separate parts. The modularity of neural architectures and their universal approximation property allow incorporation of neural components in a complex simulator with simple interfaces.

When simulators of physical processes are too complex, it is impossible to use them for onboard computation. In that case, the parsimonious approximation property of neural architectures are used to provide representations of the phenomena of interest to aid in decision making. In this kind of applications, the learning phase is performed before the operational use of the network. Neural learning is also used to tune the parameters, to solve inverse problems, and to control devices in changing environments. A major factor hindering the widespread use of neural networks in industrial applications has been the lack of tools for validation of the final results. In aeronautics and space applications, the specifications are very precise and the statistical approximations that are not directly supported by physical arguments have to be rigorously proven. Hence, in this session we emphasize research in validation of neural based tools for pattern recognition and control, especially approaches that yield practical bounds for real-world problems.

The session for aeronautics and space applications is designed to introduce relevant applications of neural architectures in a large range of problems in the area, and to allow exchange of information among researchers and system designers in the area.

For further information contact the organizer:

Prof. Manuel Samuelides
Ecole Nationale Supérieure de l'Aéronautique et de l'Espace
Département Mathématiques Appliquées
e-mail: samuelid@supaero.fr

 

SUBMISSION SCHEDULE

Submission of full paper:April 30, 2003
Notification of acceptance:May 30, 2003
Submission of photo-ready accepted paper and author registration:June 20, 2003
Advanced registration, before:July 24, 2003

 

Presentation Guideline

Oral:
The allocated time for oral presentation is 20 minutes including questions. Please limit your presentation to 17 minutes leaving 3 minutes for questions.

The lecture room will be equipped with computer and overhead projectors.

Poster:
The poster board has the dimensions: height: 180 cm, width: 80 cm Please make sure that the deployed text font size is at least 20pt.
If you need any special assistance or equipment please contact General Chair Christophe MOLINA Christophe.Molina@sanofi-synthelabo.com