62
1

DP-Dueling: Learning from Preference Feedback without Compromising User Privacy

Abstract

We consider the well-studied dueling bandit problem, where a learner aims to identify near-optimal actions using pairwise comparisons, under the constraint of differential privacy. We consider a general class of utility-based preference matrices for large (potentially unbounded) decision spaces and give the first differentially private dueling bandit algorithm for active learning with user preferences. Our proposed algorithms are computationally efficient with near-optimal performance, both in terms of the private and non-private regret bound. More precisely, we show that when the decision space is of finite size KK, our proposed algorithm yields order optimal O(i=2KlogKTΔi+Kϵ)O\Big(\sum_{i = 2}^K\log\frac{KT}{\Delta_i} + \frac{K}{\epsilon}\Big) regret bound for pure ϵ\epsilon-DP, where Δi\Delta_i denotes the suboptimality gap of the ii-th arm. We also present a matching lower bound analysis which proves the optimality of our algorithms. Finally, we extend our results to any general decision space in dd-dimensions with potentially infinite arms and design an ϵ\epsilon-DP algorithm with regret O~(d6κϵ+dTκ)\tilde{O} \left( \frac{d^6}{\kappa \epsilon } + \frac{ d\sqrt{T }}{\kappa} \right), providing privacy for free when TdT \gg d.

View on arXiv
Comments on this paper

We use cookies and other tracking technologies to improve your browsing experience on our website, to show you personalized content and targeted ads, to analyze our website traffic, and to understand where our visitors are coming from. See our policy.