Latent Contextual Bandits and their Application to Personalized Recommendations for New Users

TL;DR

This paper introduces Latent Contextual Bandits, a novel approach that improves personalized recommendations for new users by learning latent user classes, resulting in better regret bounds and practical performance.

Contribution

It proposes a new latent class-based method for contextual bandits that enhances cold-start recommendations and provides theoretical and empirical advantages over existing algorithms.

Findings

Achieves a better regret bound than existing algorithms.

Demonstrates improved recommendation quality on real-world data.

Shows benefits through a user study.

Abstract

Personalized recommendations for new users, also known as the cold-start problem, can be formulated as a contextual bandit problem. Existing contextual bandit algorithms generally rely on features alone to capture user variability. Such methods are inefficient in learning new users' interests. In this paper we propose Latent Contextual Bandits. We consider both the benefit of leveraging a set of learned latent user classes for new users, and how we can learn such latent classes from prior users. We show that our approach achieves a better regret bound than existing algorithms. We also demonstrate the benefit of our approach using a large real world dataset and a preliminary user study.