Saved in:
Bibliographic Details
Main Authors: Salgia, Sudeep, Zhao, Qing
Format: Preprint
Published: 2022
Subjects:
Online Access:https://arxiv.org/abs/2211.02212
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866914157253623808
author Salgia, Sudeep
Zhao, Qing
author_facet Salgia, Sudeep
Zhao, Qing
contents We consider distributed linear bandits where $M$ agents learn collaboratively to minimize the overall cumulative regret incurred by all agents. Information exchange is facilitated by a central server, and both the uplink and downlink communications are carried over channels with fixed capacity, which limits the amount of information that can be transmitted in each use of the channels. We investigate the regret-communication trade-off by (i) establishing information-theoretic lower bounds on the required communications (in terms of bits) for achieving a sublinear regret order; (ii) developing an efficient algorithm that achieves the minimum sublinear regret order offered by centralized learning using the minimum order of communications dictated by the information-theoretic lower bounds. For sparse linear bandits, we show a variant of the proposed algorithm offers better regret-communication trade-off by leveraging the sparsity of the problem.
format Preprint
id arxiv_https___arxiv_org_abs_2211_02212
institution arXiv
publishDate 2022
record_format arxiv
spellingShingle Distributed Linear Bandits under Communication Constraints
Salgia, Sudeep
Zhao, Qing
Machine Learning
We consider distributed linear bandits where $M$ agents learn collaboratively to minimize the overall cumulative regret incurred by all agents. Information exchange is facilitated by a central server, and both the uplink and downlink communications are carried over channels with fixed capacity, which limits the amount of information that can be transmitted in each use of the channels. We investigate the regret-communication trade-off by (i) establishing information-theoretic lower bounds on the required communications (in terms of bits) for achieving a sublinear regret order; (ii) developing an efficient algorithm that achieves the minimum sublinear regret order offered by centralized learning using the minimum order of communications dictated by the information-theoretic lower bounds. For sparse linear bandits, we show a variant of the proposed algorithm offers better regret-communication trade-off by leveraging the sparsity of the problem.
title Distributed Linear Bandits under Communication Constraints
topic Machine Learning
url https://arxiv.org/abs/2211.02212