Entire Space Counterfactual Learning for Reliable Content Recommendations
- 12 December 2024
- journal article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Information Forensics and Security
- Vol. 20 (15566013) , 1755-1764
Abstract
Post-click conversion rate (CVR) estimation is a fundamental task in developing effective recommender systems, yet it faces challenges from data sparsity and sample selection bias. To handle both challenges, employ entire space multitask models decompose the user behavior track into a sequence of exposure → click → conversion, to foster construct auxiliary learning tasks for CVR estimation. However, these methods suffer from two significant defects: (1) intrinsic estimation bias (IEB), where CVR estimates are in overestimated; (2) false independence prior (FIP), where the causal relationship between clicks and subsequent conversions might be overlooked. To overcome these limitations, we develop a model-agnostic framework, namely Entire Space Counterfactual Multitask Model (ESCM2), which incorporates a counterfactual risk minimizer within the ESMM framework to regularize CVR estimation. Experiments conducted on large-scale industrial recommendation datasets and an online industrial recommendation service demonstrate ESCM 2 effectively mitigates IEB and FIP defects and substantially enhances recommendation performance.Keywords
Funding Information
- National Natural Science Foundation of China (623B2002, 62073288, 12075212, 12105246)
This publication has 42 references indexed in Scilit:
- Toward Adversarially Robust Recommendation From Adaptive Fraudster DetectionIEEE Transactions on Information Forensics and Security, 2023
- Modeling Task Relationships in Multivariate Soft Sensor With Balanced Mixture-of-ExpertsIEEE Transactions on Industrial Informatics, 2022
- Attacking Recommender Systems With Plausible ProfileIEEE Transactions on Information Forensics and Security, 2021
- Cross-platform Item Recommendation for Online Social E-CommerceIEEE Transactions on Knowledge and Data Engineering, 2021
- Outcome-Adaptive Lasso: Variable Selection for Causal InferenceBiometrics, 2017
- Unbiased Learning-to-Rank with Biased Feedback2017
- Methods for Constructing and Assessing Propensity ScoresHealth Services Research, 2014
- Training and testing of recommender systems on data missing not at random2010
- Causality2009
- Safe and Effective Importance SamplingJournal of the American Statistical Association, 2000