Unifying Online and Counterfactual Learning to Rank: A Novel Counterfactual Estimator that Effectively Utilizes Online Interventions (Extended Abstract)
Proceedings of the Thirtieth International Joint Conference on Artificial Intelligence. Sister Conferences Best Papers. Pages 4809-4813.
https://doi.org/10.24963/ijcai.2021/656
Keywords
- Machine Learning: Recommender Systems
- Data Mining: Information Retrieval
- Machine Learning: Online Learning
Abstract
State-of-the-art Learning to Rank (LTR) methods for optimizing ranking systems based on user interactions are divided into online approaches – that learn by direct interaction – and counterfactual approaches – that learn from historical interactions. We propose a novel intervention-aware estimator to bridge this online/counterfactual division. 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 show that, unlike existing counterfactual LTR methods, the intervention-aware estimator can greatly benefit from online interventions. To the best of our knowledge, this is the first method that is shown to be highly effective in both online and counterfactual scenarios.
@inproceedings{ijcai2021p656,
title = {Unifying Online and Counterfactual Learning to Rank: A Novel Counterfactual Estimator that Effectively Utilizes Online Interventions (Extended Abstract)},
author = {Oosterhuis, Harrie and de Rijke, Maarten},
booktitle = {Proceedings of the Thirtieth International Joint Conference on
Artificial Intelligence, {IJCAI-21}},
publisher = {International Joint Conferences on Artificial Intelligence Organization},
editor = {Prof. Zhi-Hua Zhou},
pages = {4809--4813},
year = {2021},
month = {8},
note = {Sister Conferences Best Papers},
doi = {10.24963/ijcai.2021/656},
url = {https://doi.org/10.24963/ijcai.2021/656},
}
TY - CPAPER TI - Unifying Online and Counterfactual Learning to Rank: A Novel Counterfactual Estimator that Effectively Utilizes Online Interventions (Extended Abstract) AU - Oosterhuis, Harrie AU - de Rijke, Maarten T2 - Proceedings of the Thirtieth International Joint Conference on Artificial Intelligence, IJCAI-21 PB - International Joint Conferences on Artificial Intelligence Organization PY - 2021 SP - 4809 EP - 4813 DO - 10.24963/ijcai.2021/656 UR - https://doi.org/10.24963/ijcai.2021/656 ER -Copied
Machine-readable versions of this paper, for tools and AI agents:
