View all Seminars  |  Download ICal for this event

Statistics, computation and adaptation in offline and online learning

Series: Department Seminar

Speaker: Avishek Ghosh

Date/Time: Oct 12 18:30:00


Faculty Advisor:

In this talk, I will address the statistical, computational and adaptive aspects of learning theory; both in offline (batch) as well as in online settings. The first one-third of the talk deals with a computationally efficient and statistically sound Alternating Minimization (AM) algorithm (often called hard EM), typically used to solve non-convex problems. In particular, we apply AM to a classical non-convex problem, namely max-affine regression. Max-affine regression can be thought of as a generalization of the (real) Phase Retrieval problem, and closely resembles the canonical problem of convex regression in non-parametric statistics. In the next segment of the talk, I characterize the (exact) convergence speed of the AM algorithm. In particular, a super-linear convergence of AM is (theoretically) proved, resolving a long-standing (1995) conjecture of Lei Xu and Micheal I. Jordan. The final part of the talk deals with adaptation, in a non-trivial online (bandit) setting. I will talk about my recent works on model selection in contextual bandits, which partially solves an open problem of COLT 2020.

Speaker Bio:
I am currently a final year PhD student in the Electrical Engg. and Computer Sciences (EECS) department of UC Berkeley, working with Prof. Kannan Ramchandran and Prof. Aditya Guntuboyina ( and collaborate with Prof. Arya Mazumdar of UMASS, currently at Amazon Berkeley). My research interests are broadly in Theoretical machine learning, Distributed and Robust Learning, Bandits and RL. I spent the summer of 2020 working at the Supply Chain Optimization Team of Amazon Research in New York, with Alexander (Sasha) Rakhlin and Dean Foster. Before coming to Berkeley, I completed my master's degree (in ECE) from IISc Bangalore working with Prof. Anurag Kumar, and with Prof. Aditya Gopalan. Before IISc, I spent 4 years at Jadavpur University, where I completed my bachelor's degree in Electronics and Telecom Department and worked with Prof. Amit Konar. I am a recipient of the Excellency Award from the EECS Department at Berkeley, and a gold medal from Jadavpur University, among other accolades.

Host Faculty: Gugan Thoppe