Time |
Speaker |
Title |
|
8:50am-9:00am | Organizers
| Opening Remarks |
9:00am-9:45am | Leon Bottou
| TBA | [abstract] |
TBA
|
9:45am-10:30am | Yurii Nesterov
| Detecting communities by voting model | [abstract] |
In this talk, we analyze a voting model based
on random preferences of participants. Each voter can
choose a party with certain probability, which depends on
the divergence between his personal preferences and the
position of the party. Our model represents a rare example
of a community detection (or, clustering) model with
unique equilibrium solution. This is achieved by allowing
the flexible positions for the parties (centers of clusters). We
propose an efficient algorithm for finding the equilibrium
state with linear rate of convergence. It can be
interpreted as an alternating minimization scheme. Again,
this is a rare example when such a scheme has provable
global rate of convergence.
|
10:30am-11:00am | POSTER SESSION (including coffee break 10:30am - 11:00am) |
11:00am-11:15am | Yossi Arjevani, Ohad Shamir and Ron Shiff
| Oracle Complexity of Second-Order Methods for Smooth Convex Optimization |
11:15am-11:30am | Dong Yin, Ashwin Pananjady, Max Lam, Dimitris Papailiopoulos, Kannan Ramchandran and Peter Bartlett
| Gradient Diversity: a Key Ingredient for Scalable Distributed Learning |
11:30am-12:15pm | Francis Bach
| Stochastic gradient methods with exponential convergence of testing errors | [abstract] |
TBA
|
12:15pm-2:00pm | Lunch Break |
2:00pm-2:45pm | Dimitri Bertsekas
| TD(lambda) and the Proximal Algorithm | [abstract] |
We consider large linear and nonlinear fixed point problems, and solution with proximal algorithms. We show that there is a close connection between two seemingly different types of methods from distinct fields: 1) Proximal iterations for linear systems of equations, which are prominent in numerical analysis and convex optimization, and 2) Temporal difference (TD) type methods, such as TD(lambda), LSTD(lambda), and LSPE(lambda), which are central in simulation-based approximate dynamic programming/reinforcement learning (DP/RL), and its recent prominent successes in large-scale game contexts, among others.
One benefit of this connection is a new and simple way to accelerate the standard proximal algorithm by extrapolation towards the TD iteration, which generically has a faster convergence rate. Another benefit is the potential integration into the proximal algorithmic context of several new ideas that have emerged in the DP/RL context. A third benefit is that insights and analysis from proximal algorithms can be brought to bear on the enhancement of TD methods.
The linear fixed point methodology can be extended to nonlinear fixed point problems involving a contraction, thus providing guaranteed and potentially substantial acceleration of the proximal and forward backward splitting algorithms at no extra cost. Moreover, the connection of proximal and TD methods can be extended to nonlinear (nondifferentiable) fixed point problems through new proximal-like algorithms that involve successive linearization, similar to policy iteration in DP.
|
2:45pm-3:00pm | Yair Carmon, John Duchi, Oliver Hinder and Aaron Sidford
| Lower Bounds for Finding Stationary Points of Non-Convex, Smooth High-Dimensional Functions |
3:00pm-3:30pm | POSTER SESSION (including coffee break 3:00pm - 3:30pm) |
3:30pm-4:15pm | Pablo Parrilo
| TBA | [abstract] |
TBA
|
4:15pm-4:30pm | Michael Cohen, Chinmay Hegde, Stefanie Jegelka and Ludwig Schmidt
| Efficiently Optimizing over (Non-Convex) Cones via Approximate Projections |
4:30pm-6:30pm | POSTER SESSION |