Monday, August 1, 2016 - 11:00am - 12:30pm
Shuzhong Zhang (University of Minnesota, Twin Cities)
In this talk we will discuss low-complexity algorithms for solving large-scale convex optimization problems. Such algorithms include: gradient projection, proximal gradient, Iterative Shrinkage-Thresholding (ISTA), Nesterov's acceleration, and Alternating Direction Method of Multipliers (ADMM). The emphasis of the discussion will be placed on the convergence behavior of these algorithms.
Wednesday, May 20, 2015 - 2:55pm - 3:45pm
Sébastien Bubeck (Microsoft)
I will show how to use stochastic gradient descent to sample (in polynomial time) from a log-concave distribution.

Joint work with Ronen Eldan and Joseph Lehec.
Subscribe to RSS - Gradient