Unifying Online and Counterfactual Learning to Rank: A Novel Counterfactual Estimator that Effectively Utilizes Online Interventions
Authors:
Open Access
- 8 March 2021
- conference paper
- Published by Association for Computing Machinery (ACM)
- p. 463-471
Abstract
Optimizing ranking systems based on user interactions is a well-studied problem. State-of-the-art methods for optimizing ranking systems based on user interactions are divided into online approaches - that learn by directly interacting with users - and counterfactual approaches - that learn from historical interactions. Existing online methods are hindered without online interventions and thus should not be applied counterfactually. Conversely, counterfactual methods cannot directly benefit from online interventions. We propose a novel intervention-aware estimator for both counterfactual and online Learning to Rank (LTR). With the introduction of the intervention-aware estimator, we aim to bridge the online/counterfactual LTR division as it is shown to be highly effective in both online and counterfactual scenarios. The estimator corrects for the effect of position bias, trust bias, and item-selection bias by using corrections based on the behavior of the logging policy and on online interventions: changes to the logging policy made during the gathering of click data. Our experimental results, conducted in a semi-synthetic experimental setup, show that, unlike existing counterfactual LTR methods, the intervention-aware estimator can greatly benefit from online interventions.Keywords
All Related Versions
Funding Information
- Nederlandse Organisatie voor Wetenschappelijk Onderzoek (612.001.551)
This publication has 22 references indexed in Scilit:
- Unbiased Learning-to-Rank with Biased Feedback2017
- Learning to Rank with Selection Bias in Personal Search2016
- Reusing historical interaction data for faster online learning to rank for IR2013
- Novelty and Diversity in Top-N Recommendation -- Analysis and EvaluationACM Transactions on Internet Technology, 2011
- Performance of recommender algorithms on top-n recommendation tasks2010
- Learning to Rank for Information RetrievalFoundations and Trends® in Information Retrieval, 2009
- Accurately interpreting clickthrough data as implicit feedback2005
- On Real-Time Top k Querying for Mobile ServicesLecture Notes in Computer Science, 2002
- Optimizing search engines using clickthrough data2002
- The Nature of Statistical Learning Theory2000