Stochastic Dual Coordinate Descent with Bandit Sampling

12/08/2017
by   Farnood Salehi, et al.
0

Coordinate descent methods minimize a cost function by updating a single decision variable (corresponding to one coordinate) at a time. Ideally, one would update the decision variable that yields the largest marginal decrease in the cost function. However, finding this coordinate would require checking all of them, which is not computationally practical. We instead propose a new adaptive method for coordinate descent. First, we define a lower bound on the decrease of the cost function when a coordinate is updated and, instead of calculating this lower bound for all coordinates, we use a multi-armed bandit algorithm to learn which coordinates result in the largest marginal decrease while simultaneously performing coordinate descent. We show that our approach improves the convergence of the coordinate methods (including parallel versions) both theoretically and experimentally.

READ FULL TEXT

page 1

page 2

page 3

page 4

research
01/28/2020

Faster Activity and Data Detection in Massive Random Access: A Multi-armed Bandit Approach

This paper investigates the grant-free random access with massive IoT de...
research
11/03/2020

Diverging orbits for the Ehrlich-Aberth and the Weierstrass root finders

We show that the higher dimensional Weierstrass and Ehrlich-Aberth metho...
research
11/21/2017

Regret Analysis for Continuous Dueling Bandit

The dueling bandit is a learning framework wherein the feedback informat...
research
03/27/2019

Skill Acquisition via Automated Multi-Coordinate Cost Balancing

We propose a learning framework, named Multi-Coordinate Cost Balancing (...
research
12/20/2013

Accelerated, Parallel and Proximal Coordinate Descent

We propose a new stochastic coordinate descent method for minimizing the...
research
09/23/2020

Convergence of Gibbs Sampling: Coordinate Hit-and-Run Mixes Fast

The Gibbs Sampler is a general method for sampling high-dimensional dist...
research
11/13/2020

A Homotopy Coordinate Descent Optimization Method for l_0-Norm Regularized Least Square Problem

This paper proposes a homotopy coordinate descent (HCD) method to solve ...

Please sign up or login with your details

Forgot password? Click here to reset