Seminar Details
2019-10-08 (14:00) : Two 20-minute talks
At Euler building (room A.002)
Organized by Mathematical Engineering
Section 1: Proximal Method with Contractions for Smooth Convex Optimization
Speaker :
Doikov, Nikita
Abstract :
In this talk we study a construction of proximal accelerated methods for smooth convex optimization. At every step the method minimizes a sum of contracted objective and a regularizer in a form of Bregman divergence, by some auxiliary subroutine. We present complexity analysis for a general scheme. In the case when Tensor Method is used as internal method, we show accelerated rate of convergence with additional logarithmic factor as a cost for solving the subproblem.
Section 2: Greedy Quasi-Newton Method with Explicit Superlinear Convergence
Speaker :
Anton Rodomanov (CISPA, Germany)
Abstract :
We propose a new quasi-Newton method for unconstrained minimization of smooth functions. Our method is based on the famous BFGS scheme but it uses a greedily selected coordinate vector for updating the Hessian approximation instead of the previous search direction. We prove that the proposed method has local superlinear convergence and establish a precise bound for its rate. To our knowledge, this result is the first explicit non-asymptotic rate of superlinear convergence for quasi-Newton methods.