Overcoming Prior Misspecification in Online Learning to Rank

TL;DR

This paper introduces adaptive online learning to rank algorithms that effectively handle prior misspecification, extending to linear models and incorporating various feedback types, with demonstrated success on synthetic and real data.

Contribution

It proposes adaptive algorithms that overcome prior misspecification in online learning to rank, extending to linear and generalized linear models with multiple feedback types.

Findings

Algorithms perform well despite prior misspecification

Effective on synthetic and real-world datasets

Extend to scalar relevance and click feedback

Abstract

The recent literature on online learning to rank (LTR) has established the utility of prior knowledge to Bayesian ranking bandit algorithms. However, a major limitation of existing work is the requirement for the prior used by the algorithm to match the true prior. In this paper, we propose and analyze adaptive algorithms that address this issue and additionally extend these results to the linear and generalized linear models. We also consider scalar relevance feedback on top of click feedback. Moreover, we demonstrate the efficacy of our algorithms using both synthetic and real-world experiments.