Reinforcement Online Learning to Rank with Unbiased Reward Shaping

TL;DR

This paper introduces ROLTR, a reinforcement learning algorithm for online ranking that uses unbiased reward shaping to correct position bias in user click data, improving ranking performance and efficiency.

Contribution

The paper proposes a novel reinforcement learning method with unbiased reward shaping for online learning to rank, addressing click bias and modeling unclicked documents.

Findings

ROLTR achieves state-of-the-art performance on OLTR datasets.

The method reduces the amount of user interactions needed for effective training.

Empirical results show significantly better user experience compared to existing approaches.

Abstract

Online learning to rank (OLTR) aims to learn a ranker directly from implicit feedback derived from users' interactions, such as clicks. Clicks however are a biased signal: specifically, top-ranked documents are likely to attract more clicks than documents down the ranking (position bias). In this paper, we propose a novel learning algorithm for OLTR that uses reinforcement learning to optimize rankers: Reinforcement Online Learning to Rank (ROLTR). In ROLTR, the gradients of the ranker are estimated based on the rewards assigned to clicked and unclicked documents. In order to de-bias the users' position bias contained in the reward signals, we introduce unbiased reward shaping functions that exploit inverse propensity scoring for clicked and unclicked documents. The fact that our method can also model unclicked documents provides a further advantage in that less users interactions are required to effectively train a ranker, thus providing gains in efficiency. Empirical evaluation on standard OLTR datasets shows that ROLTR achieves state-of-the-art performance, and provides significantly better user experience than other OLTR approaches. To facilitate the reproducibility of our experiments, we make all experiment code available at https://github.com/ielab/OLTR.