Addressing the Data Challenge: Model Comparison with Few Samples

Friday, August 13, 2021, 11:00 am - 12:00 pm PDTiCal
Virtually via Zoom WebinarThis event is open to the public.
AI Seminar
Zhou Wang, University of Waterloo
Video Recording:

AI Seminar Series

Seminars for the Artificial Intelligence Division at USC Information Sciences Institute

YOU ONLY NEED TO REGISTER ONCE TO ATTEND THE ENTIRE SERIES - We will send you email announcements with details of the upcoming speakers.

Register in advance for this webinar:

After registering, you will receive an email confirmation containing information about joining the Zoom webinar.


Abstract: In the past decades, we have witnessed a remarkable expansion of machine learning and especially deep learning technologies in real world applications. One common challenge in a great proportion of such applications is the lack of big data, which often turns out to be essential in developing reliable machine learning models. In this talk, we will first use a few examples in the field of perceptual image quality assessment to discuss how such data challenges may be addressed. We will then focus on the problem of model comparison, where multiple models make predictions in a large domain of high dimensions, but obtaining ground truth samples are extremely expensive. Apparently, finding the best model from many in an affordable and perhaps most efficient way becomes a big challenge. We discuss the maximum differentiate (MAD) competition and group MAD (gMAD) competition methodologies we proposed in the past years, which offer an opportunity to perform model comparison with few samples. We will again use the application of image quality assessment as an example to demonstrate the ideas.


Bio: Dr. Zhou Wang is a Canada Research Chair and Professor in the Department of Electrical and Computer Engineering, University of Waterloo. He is also a co-Founder of SSIMWAVE Inc., which offers deep technology solutions to the media, entertainment and telecommunication industry. His research interests include image/video/multimedia processing, coding, communication, computational vision, and machine learning. He has more than 200 publications in these fields with over 70,000 citations by Google Scholar statistics. Dr. Wang is a Fellow of IEEE, a Fellow of Royal Society of Canada - Academy of Science, and a Fellow of Canadian Academy of Engineering. He is a recipient of 2014 Steacie Memorial Fellowship awarded by the Governor General of Canada, and several top paper awards by IEEE Signal Processing Society. He is also a two-time recipient of Technology Emmy Awards, one in 2015 as an individual, and the other in 2021 by SSIMWAVE of which he is the Chief Scientist.



Host: Mohammad Rostami, POC: Alma Nava

Speaker approved to be recorded; The recording for this AI Seminar talk will be posted on our USC/ISI AI Seminars YouTube page within 1-2 business days:

Subscribe here to learn more about upcoming seminars:

« Return to Upcoming Events