The Department of Electrical and Computer Engineering would like to extend an invitation to CEE Faculty and Students to attend our annual Lytle Lecture series, this year featuring world-renowned applied mathematician and research scientist, Dr. Stéphane Mallat. Please find information on our Tuesday, December 3 talk times and locations below.

Best,

Jessi Navarre
Department of Electrical & Computer Engineering
University of Washington
206.685.3810

Dec. 3, 2019, Lytle Lectures

Prof. Stéphane Mallat — Collège de France, Paris; Flatiron Institute, New York

The Department of Electrical & Computer Engineering’s 2019 Dean W. Lytle Endowed Lecture Series is just around the corner, and will be held on Tuesday, December 3. The Lytle Lecture Series is the Department of Electrical & Computer Engineering’s premier annual event, featuring internationally renowned researchers in the field of communications, signal processing, control systems and machine learning.

This year we are extremely excited to welcome the world-renowned applied mathematician and research scientist, Dr. Stéphane Mallat, as our guest speaker. Dr. Mallat is known for his fundamental work in wavelet theory, with major impact in machine learning, signal processing, music synthesis, harmonic analysis, and image segmentation.

Professor Mallat will be giving two lectures on December 3, 2019. Both are free and open to the public:

Lecture #1: Technical Colloquium, open to the public
Tuesday, Dec. 3, 10:30 to 11:30 a.m., ECE Building Room 105, University of Washington
“Interpretable Deep Networks for Classification, Generation and Physics“

Abstract: Approximating high-dimensional functionals with low-dimensional models is a central issue of machine learning, image processing, physics and mathematics. Deep convolutional networks are able to approximate such functionals over a wide range of applications. This talk shows that these computational architectures take advantage of scale separation, symmetries and sparse representations. We introduce simplified architectures which can be analyzed mathematically. Scale separation is performed with wavelets and scale interactions are captured through phase coherence. Pattern structures are captured by sparse coding in dictionaries. We show applications to image classification and generation, as well as regression of quantum molecular energies and modelization of turbulent flows.

Lecture #2: For broad audience, open to the public
Tuesday, Dec. 3, 3:30 to 5:30 p.m. (doors open at 3:00 p.m.), Paul G. Allen Center Atrium, University of Washington
“Mathematical Mysteries of Deep Neural Networks”

To reserve your seat for this event, please click on the registration button below.

Abstract: Deep neural networks obtain impressive results for image, sound and language recognition or to address complex problems in physics. They are partly responsible for the renewal of artificial intelligence. Yet, we do not understand why they can work so well and why they sometimes fail, which raises many problems of robustness and explainability.

Recognizing or classifying data amounts to approximate phenomena which depend on a very large number of variables. The combinatorial explosion of possibilities makes it potentially impossible to solve. One can learn from data only if the problem is highly structured. Deep neural networks appear to take advantage of the existence of such structures, whose nature seem to be similar across a wide range of applications. Understanding this “architecture of complexity” involves many branches of mathematics and is related to open questions in physics. I will discuss some approaches and show applications.

The Lytle Lecture Series is hosted by UW ECE professors Les Atlas (atlas@uw.edu) and Maryam Fazel (mfazel@uw.edu). For more information on these and previous Lytle Lecture Series events, please visit our website. We look forward to seeing you!

Best regards,

Events & Promotions Team
UW Electrical & Computer Engineering
events@ece.uw.edu