Scalar is Not Enough: Vectorization-based Unbiased Learning to Rank

Abstract

Unbiased learning to rank (ULTR) aims to train an unbiased ranking model from biased user click logs. Most of the current ULTR methods are based on the examination hypothesis (EH), which assumes that the click probability can be factorized into two scalar functions, one related to ranking features and the other related to bias factors. Unfortunately, the interactions among features, bias factors and clicks are complicated in practice, and usually cannot be factorized in this independent way. Fitting click data with EH could lead to model misspecification and bring the approximation error. In this paper, we propose a vector-based EH and formulate the click probability as a dot product of two vector functions. This solution is complete due to its universality in fitting arbitrary click functions. Based on it, we propose a novel model named Vectorization to adaptively learn the relevance embeddings and sort documents by projecting embeddings onto a base vector. Extensive experiments show that our method significantly outperforms the state-of-the-art ULTR methods on complex real clicks as well as simple simulated clicks.

Accepted by KDD 2022

References in corpus (9)

• What Uncertainties Do We Need in Bayesian Deep Learning for Computer Vision?
• Unbiased Learning to Rank with Unbiased Propensity Estimation
• Correcting for Selection Bias in Learning-to-rank Systems
• Estimating Position Bias without Intrusive Interventions
• Intervention Harvesting for Context-Dependent Examination-Bias Estimation
• When Inverse Propensity Scoring does not Work: Affine Corrections for Unbiased Learning to Rank
• Cascade Model-based Propensity Estimation for Counterfactual Learning to Rank
• Unbiased Learning to Rank: Online or Offline?
• Counterfactual Learning to Rank using Heterogeneous Treatment Effect Estimation