A Random Walk Approach to First-Order Stochastic Convex Optimization

01/17/2019
by   Sattar Vakili, et al.
0

Online minimization of an unknown convex function over a convex and compact set is considered under first-order stochastic bandit feedback, which returns a random realization of the gradient of the function at each query point. Without knowing the distribution of the random gradients, a learning algorithm sequentially chooses query points with the objective of minimizing regret defined as the expected cumulative loss of the function values at the query points in excess to the minimum value of the function. An active search strategy based on devising a biased random walk on an infinite-depth tree constructed through successive partitioning of the domain of the function is developed. It is shown that the biased random walk moves toward the optimal point in a geometric rate, leading to an order-optimal regret performance of O(√(T)). The structural properties of this random-walk based strategy admits detailed finite-time regret analysis. By localizing data processing to small subsets of the input domain based on the tree structure, it enjoys O(1) computation and memory complexity per query and allows dynamic allocation of limited data storage.

READ FULL TEXT

page 1

page 2

page 3

page 4

research
02/11/2014

On Zeroth-Order Stochastic Convex Optimization via Random Walks

We propose a method for zeroth order stochastic convex optimization that...
research
10/04/2020

"Drunk Man" Saves Our Lives: Route Planning by a Biased Random Walk Mode

Based on the hurricane striking Puerto Rico in 2017, we developed a tran...
research
05/22/2019

Heavy Hitters and Bernoulli Convolutions

A very simple event frequency approximation algorithm that is sensitive ...
research
11/03/2020

Random Walk Bandits

Bandit learning problems find important applications ranging from medica...
research
06/25/2021

Jitter: Random Jittering Loss Function

Regularization plays a vital role in machine learning optimization. One ...
research
09/25/2021

Random Walk-steered Majority Undersampling

In this work, we propose Random Walk-steered Majority Undersampling (RWM...
research
12/11/2018

Distribution-free properties of isotonic regression

It is well known that the isotonic least squares estimator is characteri...

Please sign up or login with your details

Forgot password? Click here to reset