Deep Exploration for Recommendation Systems

TL;DR

This paper introduces deep exploration methods for recommendation systems, framing them as sequential decision problems to effectively learn from delayed and sparse feedback, leading to significant performance improvements.

Contribution

It formulates recommendation as a sequential decision process and develops deep exploration techniques to better learn from delayed feedback, especially in sparse reward scenarios.

Findings

Deep exploration outperforms single-step exploration.

Significant improvements demonstrated in industrial-grade simulators.

Effective learning from delayed and sparse feedback achieved.

Abstract

Modern recommendation systems ought to benefit by probing for and learning from delayed feedback. Research has tended to focus on learning from a user's response to a single recommendation. Such work, which leverages methods of supervised and bandit learning, forgoes learning from the user's subsequent behavior. Where past work has aimed to learn from subsequent behavior, there has been a lack of effective methods for probing to elicit informative delayed feedback. Effective exploration through probing for delayed feedback becomes particularly challenging when rewards are sparse. To address this, we develop deep exploration methods for recommendation systems. In particular, we formulate recommendation as a sequential decision problem and demonstrate benefits of deep exploration over single-step exploration. Our experiments are carried out with high-fidelity industrial-grade simulators and establish large improvements over existing algorithms.