Deep Network Representations for High Dimensional Signal Analysis
Event details
| Date | 25.08.2014 |
| Hour | 11:15 |
| Speaker |
Prof. Stéphane Mallat, Ecole normale supérieure, Paris Bio: Stéphane Mallat received the Ph.D. degree in electrical engineering from the University of Pennsylvania, Philadelphia, in 1988. In 1988, he joined the Computer Science Department of the Courant Institute of Mathematical Sciences where he was Associate Professor in 1994 and Professor in 1996. Since 1995, he has been a full Professor in the Applied Mathematics Department at Ecole Polytechnique, Paris. From 2001 to 2008 he was a co-founder and CEO of a start-up company. Prof. Mallat is an IEEE and EURASIP fellow. He received the 1990 IEEE Signal Processing Society's paper award, the 1993 Alfred Sloan fellowship in Mathematics, the 1997 Outstanding Achievement Award from the SPIE Optical Engineering Society, the 1997 Blaise Pascal Prize in applied mathematics from the French Academy of Sciences, the 2004 European IST Grand prize, the 2004 INIST-CNRS prize for most cited French researcher in engineering and computer science, and the 2007 EADS prize of the French Academy of Sciences. His research interests include signal processing, learning and harmonic analysis. |
| Location | |
| Category | Conferences - Seminars |
Remarkable signal analysis results have been obtained with deep convolution neural networks, through complex and huge optimization algorithms which are difficult to understand. This talks aims at better understanding these networks, and show that most of it may not require any learning for structured signals such as image or audio signals. Similar results are obtained a non-linear filter bank algorithm, called a scattering transform, having the same architecture but using predefined wavelet filters. Applications to various audio and image classification problems will be shown, as well the regression of molecule energies in quantum chemistry.
Practical information
- General public
- Free
Organizer
- Juri Ranieri
Contact
- Juri Ranieri