Learning to Counterfactually Explain Recommendations

TL;DR

This paper introduces a learning-based framework for generating counterfactual explanations in recommender systems, making explanations more understandable, verifiable, and user-controllable by efficiently approximating the effects of user history removal.

Contribution

The paper proposes a surrogate model approach to efficiently generate counterfactual explanations without retraining recommender models repeatedly, improving explanation validity and user satisfaction.

Findings

Outperforms baselines in counterfactual validity

Produces more satisfactory explanations in user studies

Reduces computational cost of generating explanations

Abstract

Recommender system practitioners are facing increasing pressure to explain recommendations. We explore how to explain recommendations using counterfactual logic, i.e. "Had you not interacted with the following items, we would not recommend it." Compared to the traditional explanation logic, counterfactual explanations are easier to understand, more technically verifiable, and more informative in terms of giving users control over recommendations. The major challenge of generating such explanations is the computational cost because it requires repeatedly retraining the models to obtain the effect on a recommendation caused by the absence of user history. We propose a learning-based framework to generate counterfactual explanations. The key idea is to train a surrogate model to learn the effect of removing a subset of user history on the recommendation. To this end, we first artificially simulate the counterfactual outcomes on the recommendation after deleting subsets of history. Then we train a surrogate model to learn the mapping between a history deletion and the corresponding change of the recommendation caused by the deletion. Finally, to generate an explanation, we find the history subset predicted by the surrogate model that is most likely to remove the recommendation. Through offline experiments and online user studies, we show our method, compared to baselines, can generate explanations that are more counterfactually valid and more satisfactory considered by users.