Talks & Seminars
Title: A Saddle Point Approach to Structured Low-Rank Matrix Learning
Dr. Pratik Jawanpuria, Amazon
Date & Time: August 23, 2017 14:30
Venue: Lecture Hall, 03rd Floor, B Block, Department of Computer Science and Engineering, Kanwal Rekhi (KReSIT) Building
Learning low-rank matrices is an important problem, having several applications such as recommender systems (Netflix problem), multivariate regression, etc. The structural constraints such as non-negativity, bounds in form of inequality constraints, Hankel constraints, probability constraints, or constraints due to robust loss functions are very common in applications such as image completion, system identification, among others. In this talk, we will first discuss the existing works in this area. Then I will present our novel and unified framework to solve such problems. Keywords: saddle point optimization, duality, matrix factorization. Arxiv link to the paper: https://arxiv.org/abs/1704.07352 Link to the code: https://pratikjawanpuria.com/publications/
Speaker Profile:
Pratik Jawanpuria is an applied scientist at Amazon. He belongs to the Core Machine Learning group. He completed his B.Tech in Computer Science and Ph. D. from IIT Bombay. He then worked as a postdoctoral researcher at Saarland University. His research interests lie in the areas of machine learning and optimization.
List of Talks


Faculty CSE IT
Forgot Password
    [+] Sitemap     Feedback