BLISS Seminar

About

The BLISS seminar (formerly NCD seminar) is co-sponsored by generous grants from Microsoft Research and Qualcomm, and is the area seminar of the Berkeley Laboratory for Information Systems and Sciences. Talks at the seminar cover topics including but not limited to information and coding theory, signal processing, optimization and statistics. The list of talks for the current semester can be found below, and past seminars from 2016 onwards are listed here. For an archive of all talks from 1996-2015, visit the old webpage.

Fall 2017
Location: 540 Cory Hall
Regular seminar time: Monday 3-4PM

To subscribe to our mailing list, click here.

To give a talk at the seminar, contact Tom Courtade, Ashwin Pananjady, or Orhan Ocal.

Fall 2017 Talks

Dates marked in bold indicate that talks are at non-regular times.

Sep 25 Stephen Wright (UW Madison) Algorithmic Tools for Smooth Nonconvex Optimization details
Oct 2 Stefanie Jegelka (MIT) Submodularity and Optimal Transport in ML details
Oct 4 Gerald Friedland (Berkeley) Capacity Scaling of Neural Networks details
Oct 9 Zeyuan Allen-Zhu (Microsoft Research) Optimal Experimental Design via Regret Minimization details
Oct 16 Yin Tat Lee (U. Washington) Kannan-Lovasz-Simonovitz (KLS) conjecture details
Oct 23 Sebastien Bubeck (Microsoft Research) Sparsity, variance, and curvature in multi-armed bandits details
Oct 30 Ludwig Schmidt (MIT) Efficiently Optimizing over (Non-Convex) Cones Using Approximate Projections details
Nov 6 Jiantao Jiao (Stanford) Modern Estimation of Information Theoretic Functionals details
Nov 8 Mina Karzand (MIT) Regret bounds for a latent variable recommendation systems model details
Nov 13 Lieven Vandenberghe (UCLA) Primal-dual first-order methods for convex optimization details
Nov 20 Chi Jin (Berkeley)
Nov 22 Shusen Wang (Berkeley) Communication-Efficient Distributed Optimization via a 2nd-Order Method details
Nov 27 Philipp Walk (Caltech)
Dec 11 Marco Mondelli (Stanford)