An Optimal Private Stochastic-MAB Algorithm Based on an Optimal Private Stopping Rule

05/22/2019
by   Touqir Sajed, et al.
0

We present a provably optimal differentially private algorithm for the stochastic multi-arm bandit problem, as opposed to the private analogue of the UCB-algorithm [Mishra and Thakurta, 2015; Tossou and Dimitrakakis, 2016] which doesn't meet the recently discovered lower-bound of Ω(K(T)/ϵ) [Shariff and Sheffet, 2018]. Our construction is based on a different algorithm, Successive Elimination [Even-Dar et al. 2002], that repeatedly pulls all remaining arms until an arm is found to be suboptimal and is then eliminated. In order to devise a private analogue of Successive Elimination we visit the problem of private stopping rule, that takes as input a stream of i.i.d samples from an unknown distribution and returns a multiplicative (1 ±α)-approximation of the distribution's mean, and prove the optimality of our private stopping rule. We then present the private Successive Elimination algorithm which meets both the non-private lower bound [Lai and Robbins, 1985] and the above-mentioned private lower bound. We also compare empirically the performance of our algorithm with the private UCB algorithm.

READ FULL TEXT

page 1

page 2

page 3

page 4

research
06/11/2020

Best-Arm Identification for Quantile Bandits with Privacy

We study the best-arm identification problem in multi-armed bandits with...
research
02/16/2021

Optimal Algorithms for Private Online Learning in a Stochastic Environment

We consider two variants of private stochastic online learning. The firs...
research
03/23/2021

Bandits with many optimal arms

We consider a stochastic bandit problem with a possibly infinite number ...
research
02/09/2022

Optimal Clustering with Bandit Feedback

This paper considers the problem of online clustering with bandit feedba...
research
06/04/2021

Optimal Rates of (Locally) Differentially Private Heavy-tailed Multi-Armed Bandits

In this paper we study the problem of stochastic multi-armed bandits (MA...
research
01/31/2023

Differentially-Private Hierarchical Clustering with Provable Approximation Guarantees

Hierarchical Clustering is a popular unsupervised machine learning metho...
research
04/19/2020

An Algorithm for the Exact Treedepth Problem

We present a novel algorithm for the minimum-depth elimination tree prob...

Please sign up or login with your details

Forgot password? Click here to reset