Unbiased Learning to Rank with Unbiased Propensity Estimation
arXiv:1804.05938 · doi:10.1145/3209978.3209986
Abstract
Learning to rank with biased click data is a well-known challenge. A variety of methods has been explored to debias click data for learning to rank such as click models, result interleaving and, more recently, the unbiased learning-to-rank framework based on inverse propensity weighting. Despite their differences, most existing studies separate the estimation of click bias (namely the \textit{propensity model}) from the learning of ranking algorithms. To estimate click propensities, they either conduct online result randomization, which can negatively affect the user experience, or offline parameter estimation, which has special requirements for click data and is optimized for objectives (e.g. click likelihood) that are not directly related to the ranking performance of the system. In this work, we address those problems by unifying the learning of propensity models and ranking models. We find that the problem of estimating a propensity model from click data is a dual problem of unbiased learning to rank. Based on this observation, we propose a Dual Learning Algorithm (DLA) that jointly learns an unbiased ranker and an \textit{unbiased propensity model}. DLA is an automatic unbiased learning-to-rank framework as it directly learns unbiased ranking models from biased click data without any preprocessing. It can adapt to the change of bias distributions and is applicable to online learning. Our empirical experiments with synthetic and real-world data show that the models trained with DLA significantly outperformed the unbiased learning-to-rank algorithms based on result randomization and the models trained with relevance signals extracted by click models.
References in corpus (3)
Cited by in corpus (58)
- Causal Intervention for Leveraging Popularity Bias in Recommendation
- Controlling Fairness and Bias in Dynamic Learning-to-Rank
- Learning a Deep Listwise Context Model for Ranking Refinement
- Correcting for Selection Bias in Learning-to-rank Systems
- Estimating Position Bias without Intrusive Interventions
- Unifying Online and Counterfactual Learning to Rank
- To Model or to Intervene: A Comparison of Counterfactual and Online Learning to Rank from User Interactions
- Policy-Aware Unbiased Learning to Rank for Top-k Rankings
- Intervention Harvesting for Context-Dependent Examination-Bias Estimation
- Unbiased Learning for the Causal Effect of Recommendation
- Deconfounded Causal Collaborative Filtering
- Maximizing Marginal Fairness for Dynamic Learning to Rank
- Non-Clicks Mean Irrelevant? Propensity Ratio Scoring As a Correction
- Uncovering User Interest from Biased and Noised Watch Time in Video Recommendation
- Deep Position-wise Interaction Network for CTR Prediction
- Taking the Counterfactual Online: Efficient and Unbiased Online Evaluation for Ranking
- Reaching the End of Unbiasedness: Uncovering Implicit Limitations of Click-Based Learning to Rank
- Fairness of Exposure in Light of Incomplete Exposure Estimation
- A Probabilistic Position Bias Model for Short-Video Recommendation Feeds
- Mixture-Based Correction for Position and Trust Bias in Counterfactual Learning to Rank
- Contrastive Learning for Debiased Candidate Generation in Large-Scale Recommender Systems
- The Role of Relevance in Fair Ranking
- Unbiased Learning to Rank with Biased Continuous Feedback
- Position Bias Estimation for Unbiased Learning-to-Rank in eCommerce Search
- Unbiased Learning to Rank: Online or Offline?
- Unbiased Pairwise Learning from Implicit Feedback for Recommender Systems without Biased Variance Control
- Marginal-Certainty-aware Fair Ranking Algorithm
- Vertical Allocation-based Fair Exposure Amortizing in Ranking
- Analysis of Multivariate Scoring Functions for Automatic Unbiased Learning to Rank
- Learning More From Less: Towards Strengthening Weak Supervision for Ad-Hoc Retrieval
- Learning from User Interactions with Rankings: A Unification of the Field
- On the Impact of Outlier Bias on User Clicks
- DebCSE: Rethinking Unsupervised Contrastive Sentence Embedding Learning in the Debiasing Perspective
- Click-Conversion Multi-Task Model with Position Bias Mitigation for Sponsored Search in eCommerce
- Automatic Feature Fairness in Recommendation via Adversaries
- Approximated Doubly Robust Search Relevance Estimation
- Scalar is Not Enough: Vectorization-based Unbiased Learning to Rank
- FARA: Future-aware Ranking Algorithm for Fairness Optimization
- Offline Comparison of Ranking Functions using Randomized Data
- A Conceptual Framework for Evaluating Fairness in Search
- Handling Position Bias for Unbiased Learning to Rank in Hotels Search
- CausCF: Causal Collaborative Filtering for RecommendationEffect Estimation
- Dual Learning Algorithm for Delayed Conversions
- A General Framework for Counterfactual Learning-to-Rank
- CWRCzech: 100M Query-Document Czech Click Dataset and Its Application to Web Relevance Ranking
- Learning Robust Recommender from Noisy Implicit Feedback
- Rethinking Click Models in Light of Carousel Interfaces: Theory-Based Categorization and Design of Click Models
- A Deep Recurrent Survival Model for Unbiased Ranking
- Investigating the Robustness of Counterfactual Learning to Rank Models: A Reproducibility Study
- Counterfactual Learning to Rank using Heterogeneous Treatment Effect Estimation
- Optimizing Ranking Models in an Online Setting
- An Epistemic Position-Based Click Model: From Interactions to Epistemic Distributions of Relevance and Bias
- Unbiased Learning to Rank with Query-Level Click Propensity Estimation: Beyond Pointwise Observation and Relevance
- Unbiased Pairwise Learning to Rank in Recommender Systems
- Ranker-agnostic Contextual Position Bias Estimation
- ULTRA: An Unbiased Learning To Rank Algorithm Toolbox
- Large-scale Causal Approaches to Debiasing Post-click Conversion Rate Estimation with Multi-task Learning
- Searching Personal Collections