Recommending with Recommendations

TL;DR

This paper presents a novel recommendation framework that leverages recommendations from other services to protect user privacy, using a multi-armed bandit approach with theoretical guarantees and empirical validation.

Contribution

It introduces a source-based recommendation method using a latent space decomposition and an explore-then-refine algorithm with regret guarantees.

Findings

Achieves regret comparable to full knowledge skyline.

Effectively decomposes user information into systematic and idiosyncratic components.

Outperforms benchmarks on synthetic data.

Abstract

Recommendation systems are a key modern application of machine learning, but they have the downside that they often draw upon sensitive user information in making their predictions. We show how to address this deficiency by basing a service's recommendation engine upon recommendations from other existing services, which contain no sensitive information by nature. Specifically, we introduce a contextual multi-armed bandit recommendation framework where the agent has access to recommendations for other services. In our setting, the user's (potentially sensitive) information belongs to a high-dimensional latent space, and the ideal recommendations for the source and target tasks (which are non-sensitive) are given by unknown linear transformations of the user information. So long as the tasks rely on similar segments of the user information, we can decompose the target recommendation problem into systematic components that can be derived from the source recommendations, and idiosyncratic components that are user-specific and cannot be derived from the source, but have significantly lower dimensionality. We propose an explore-then-refine approach to learning and utilizing this decomposition; then using ideas from perturbation theory and statistical concentration of measure, we prove our algorithm achieves regret comparable to a strong skyline that has full knowledge of the source and target transformations. We also consider a generalization of our algorithm to a model with many simultaneous targets and no source. Our methods obtain superior empirical results on synthetic benchmarks.