Differentially Private Online Submodular Optimization

TL;DR

This paper introduces the first differentially private algorithms for online submodular minimization, achieving low regret in both full information and bandit feedback settings by leveraging convex relaxations and estimation techniques.

Contribution

It develops novel algorithms for differentially private online submodular minimization under full information and bandit feedback, with theoretical regret guarantees.

Findings

Achieves $ ilde{O}(n^{3/2}rac{ oot{2} ext{T}}{ ext{epsilon}})$ regret in full information setting.

Achieves $ ilde{O}(n^{3/2}T^{3/4}/ ext{epsilon})$ regret in bandit setting.

First algorithms to ensure differential privacy in online submodular optimization.

Abstract

In this paper we develop the first algorithms for online submodular minimization that preserve differential privacy under full information feedback and bandit feedback. A sequence of $T$ submodular functions over a collection of $n$ elements arrive online, and at each timestep the algorithm must choose a subset of $[n]$ before seeing the function. The algorithm incurs a cost equal to the function evaluated on the chosen set, and seeks to choose a sequence of sets that achieves low expected regret. Our first result is in the full information setting, where the algorithm can observe the entire function after making its decision at each timestep. We give an algorithm in this setting that is $\epsilon$-differentially private and achieves expected regret $\tilde{O}\left(\frac{n^{3/2}\sqrt{T}}{\epsilon}\right)$. This algorithm works by relaxing submodular function to a convex function using the Lovasz extension, and then simulating an algorithm for differentially private online convex optimization. Our second result is in the bandit setting, where the algorithm can only see the cost incurred by its chosen set, and does not have access to the entire function. This setting is significantly more challenging because the algorithm does not receive enough information to compute the Lovasz extension or its subgradients. Instead, we construct an unbiased estimate using a single-point estimation, and then simulate private online convex optimization using this estimate. Our algorithm using bandit feedback is $\epsilon$-differentially private and achieves expected regret $\tilde{O}\left(\frac{n^{3/2}T^{3/4}}{\epsilon}\right)$.