Differentially Private Stochastic Linear Bandits: (Almost) for Free

07/07/2022
by   Osama A. Hanna, et al.
0

In this paper, we propose differentially private algorithms for the problem of stochastic linear bandits in the central, local and shuffled models. In the central model, we achieve almost the same regret as the optimal non-private algorithms, which means we get privacy for free. In particular, we achieve a regret of Õ(√(T)+1/ϵ) matching the known lower bound for private linear bandits, while the best previously known algorithm achieves Õ(1/ϵ√(T)). In the local case, we achieve a regret of Õ(1/ϵ√(T)) which matches the non-private regret for constant ϵ, but suffers a regret penalty when ϵ is small. In the shuffled model, we also achieve regret of Õ(√(T)+1/ϵ) central case, while the best previously known algorithm suffers a regret of Õ(1/ϵT^3/5). Our numerical evaluation validates our theoretical results.

READ FULL TEXT

page 1

page 2

page 3

page 4

research
01/27/2017

The Price of Differential Privacy For Online Learning

We design differentially private algorithms for the problem of online li...
research
06/01/2020

Locally Differentially Private (Contextual) Bandits Learning

We study locally differentially private (LDP) bandits learning in this p...
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
06/06/2018

Mitigating Bias in Adaptive Data Gathering via Differential Privacy

Data that is gathered adaptively --- via bandit algorithms, for example ...
research
07/12/2022

Differentially Private Linear Bandits with Partial Distributed Feedback

In this paper, we study the problem of global reward maximization with o...
research
04/23/2023

Robust and differentially private stochastic linear bandits

In this paper, we study the stochastic linear bandit problem under the a...
research
03/01/2023

Subset-Based Instance Optimality in Private Estimation

We propose a new definition of instance optimality for differentially pr...

Please sign up or login with your details

Forgot password? Click here to reset