Anonymous Learning via Look-Alike Clustering: A Precise Analysis of Model Generalization
This work addresses privacy concerns in personalized recommendation systems by providing a theoretical analysis of anonymous learning, though it is incremental as it builds on existing clustering techniques.
The paper tackles the problem of training models with anonymous data for privacy in recommendation systems by analyzing look-alike clustering, showing that in high-dimensional regimes, this method acts as regularization and improves generalization error, with finite-sample experiments matching theory for sample sizes of a few hundred.
While personalized recommendations systems have become increasingly popular, ensuring user data protection remains a top concern in the development of these learning systems. A common approach to enhancing privacy involves training models using anonymous data rather than individual data. In this paper, we explore a natural technique called \emph{look-alike clustering}, which involves replacing sensitive features of individuals with the cluster's average values. We provide a precise analysis of how training models using anonymous cluster centers affects their generalization capabilities. We focus on an asymptotic regime where the size of the training set grows in proportion to the features dimension. Our analysis is based on the Convex Gaussian Minimax Theorem (CGMT) and allows us to theoretically understand the role of different model components on the generalization error. In addition, we demonstrate that in certain high-dimensional regimes, training over anonymous cluster centers acts as a regularization and improves generalization error of the trained models. Finally, we corroborate our asymptotic theory with finite-sample numerical experiments where we observe a perfect match when the sample size is only of order of a few hundreds.