Characterizing the Sample Complexity of Private Learners

TL;DR

This paper provides a combinatorial characterization of the sample complexity required for private learning, introducing the notion of probabilistic representation and the RepDim measure, analogous to VC dimension in non-private learning.

Contribution

It introduces the concept of probabilistic representation and RepDim, establishing a tight characterization of the sample complexity for private learning.

Findings

RepDim is both necessary and sufficient for private learning sample complexity

The characterization parallels VC dimension for non-private learning

Applicable to private data release and optimization problems

Abstract

In 2008, Kasiviswanathan et al. defined private learning as a combination of PAC learning and differential privacy. Informally, a private learner is applied to a collection of labeled individual information and outputs a hypothesis while preserving the privacy of each individual. Kasiviswanathan et al. gave a generic construction of private learners for (finite) concept classes, with sample complexity logarithmic in the size of the concept class. This sample complexity is higher than what is needed for non-private learners, hence leaving open the possibility that the sample complexity of private learning may be sometimes significantly higher than that of non-private learning. We give a combinatorial characterization of the sample size sufficient and necessary to privately learn a class of concepts. This characterization is analogous to the well known characterization of the sample complexity of non-private learning in terms of the VC dimension of the concept class. We introduce the notion of probabilistic representation of a concept class, and our new complexity measure RepDim corresponds to the size of the smallest probabilistic representation of the concept class. We show that any private learning algorithm for a concept class C with sample complexity m implies RepDim(C)=O(m), and that there exists a private learning algorithm with sample complexity m=O(RepDim(C)). We further demonstrate that a similar characterization holds for the database size needed for privately computing a large class of optimization problems and also for the well studied problem of private data release.