Handling Cold-Start Collaborative Filtering with Reinforcement Learning

TL;DR

This paper introduces a reinforcement learning approach using Deep Q Networks to optimize interview questions for cold-start users in recommender systems, aiming to improve user profiling and recommendation quality.

Contribution

It presents a novel method for learning effective interview questions for cold-start users using Deep Q Networks, which can generalize across different recommender systems.

Findings

Improved user profiling for cold-start users.

Effective question sequences learned via reinforcement learning.

Potential for generalization across recommender domains.

Abstract

A major challenge in recommender systems is handling new users, whom are also called $\textit{cold-start}$ users. In this paper, we propose a novel approach for learning an optimal series of questions with which to interview cold-start users for movie recommender systems. We propose learning interview questions using Deep Q Networks to create user profiles to make better recommendations to cold-start users. While our proposed system is trained using a movie recommender system, our Deep Q Network model should generalize across various types of recommender systems.