What If, But Privately: Private Counterfactual Retrieval
This work addresses privacy concerns for users in high-stake applications where counterfactual explanations are needed, offering incremental improvements in privacy-preserving retrieval methods.
The paper tackles the problem of retrieving counterfactual explanations from a database while ensuring perfect, information-theoretic privacy for the user's feature vector, and proposes schemes that also reduce database leakage and handle immutable features and user preferences, with numerical results comparing database leakage.
Transparency and explainability are two important aspects to be considered when employing black-box machine learning models in high-stake applications. Providing counterfactual explanations is one way of catering this requirement. However, this also poses a threat to the privacy of the institution that is providing the explanation, as well as the user who is requesting it. In this work, we are primarily concerned with the user's privacy who wants to retrieve a counterfactual instance, without revealing their feature vector to the institution. Our framework retrieves the exact nearest neighbor counterfactual explanation from a database of accepted points while achieving perfect, information-theoretic, privacy for the user. First, we introduce the problem of private counterfactual retrieval (PCR) and propose a baseline PCR scheme that keeps the user's feature vector information-theoretically private from the institution. Building on this, we propose two other schemes that reduce the amount of information leaked about the institution database to the user, compared to the baseline scheme. Second, we relax the assumption of mutability of all features, and consider the setting of immutable PCR (I-PCR). Here, the user retrieves the nearest counterfactual without altering a private subset of their features, which constitutes the immutable set, while keeping their feature vector and immutable set private from the institution. For this, we propose two schemes that preserve the user's privacy information-theoretically, but ensure varying degrees of database privacy. Third, we extend our PCR and I-PCR schemes to incorporate user's preference on transforming their attributes, so that a more actionable explanation can be received. Finally, we present numerical results to support our theoretical findings, and compare the database leakage of the proposed schemes.