Information Theoretic Counterfactual Learning from Missing-Not-At-Random Feedback

TL;DR

This paper introduces an information-theoretic counterfactual variational information bottleneck (CVIB) to improve debiasing in recommendation systems dealing with missing-not-at-random data, without relying on costly randomized controlled trials.

Contribution

It proposes a novel CVIB method that separates factual and counterfactual information, enabling effective debiasing without RCTs in recommendation systems.

Findings

CVIB significantly improves model performance on real-world datasets.

The method facilitates balanced learning between factual and counterfactual domains.

Empirical results demonstrate the effectiveness of CVIB in counterfactual learning.

Abstract

Counterfactual learning for dealing with missing-not-at-random data (MNAR) is an intriguing topic in the recommendation literature since MNAR data are ubiquitous in modern recommender systems. Missing-at-random (MAR) data, namely randomized controlled trials (RCTs), are usually required by most previous counterfactual learning methods for debiasing learning. However, the execution of RCTs is extraordinarily expensive in practice. To circumvent the use of RCTs, we build an information-theoretic counterfactual variational information bottleneck (CVIB), as an alternative for debiasing learning without RCTs. By separating the task-aware mutual information term in the original information bottleneck Lagrangian into factual and counterfactual parts, we derive a contrastive information loss and an additional output confidence penalty, which facilitates balanced learning between the factual and counterfactual domains. Empirical evaluation on real-world datasets shows that our CVIB significantly enhances both shallow and deep models, which sheds light on counterfactual learning in recommendation that goes beyond RCTs.