Multi-Task Differential Privacy Under Distribution Skew

TL;DR

This paper addresses multi-task learning with user-level differential privacy, proposing adaptive algorithms to handle task distribution skew, which improves utility especially for tasks with fewer data samples, validated by experiments on recommendation benchmarks.

Contribution

It introduces a systematic analysis and a generic adaptive reweighting algorithm for privacy budget allocation in skewed multi-task settings.

Findings

Adaptive reweighting improves empirical risk in skewed tasks.

Method achieves state-of-the-art results on recommendation benchmarks.

Algorithms effectively handle distribution skew for better utility.

Abstract

We study the problem of multi-task learning under user-level differential privacy, in which $n$ users contribute data to $m$ tasks, each involving a subset of users. One important aspect of the problem, that can significantly impact quality, is the distribution skew among tasks. Certain tasks may have much fewer data samples than others, making them more susceptible to the noise added for privacy. It is natural to ask whether algorithms can adapt to this skew to improve the overall utility. We give a systematic analysis of the problem, by studying how to optimally allocate a user's privacy budget among tasks. We propose a generic algorithm, based on an adaptive reweighting of the empirical loss, and show that when there is task distribution skew, this gives a quantifiable improvement of excess empirical risk. Experimental studies on recommendation problems that exhibit a long tail of small tasks, demonstrate that our methods significantly improve utility, achieving the state of the art on two standard benchmarks.