PAR: Plausibility-aware Amortized Recourse Generation

TL;DR

PAR introduces a probabilistic, efficient method for generating plausible, actionable recourses that are tailored to individual facts, improving over existing approaches in validity, similarity, and plausibility.

Contribution

It formulates recourse as a probabilistic inference problem and proposes PAR, an amortized inference method that efficiently generates highly likely, realistic recourses with a neighborhood-based conditioning mechanism.

Findings

PAR outperforms existing methods in validity and plausibility of recourses.

Recourses generated by PAR are highly similar to the factual and sparse.

The method is computationally efficient and scalable.

Abstract

Algorithmic recourse aims to recommend actionable changes to a factual's attributes that flip an unfavorable model decision while remaining realistic and feasible. We formulate recourse as a Constrained Maximum A-Posteriori (MAP) inference problem under the accepted-class data distribution seeking counterfactuals with high likelihood while respecting other recourse constraints. We present PAR, an amortized approximate inference procedure that generates highly likely recourses efficiently. Recourse likelihood is estimated directly using tractable probabilistic models that admit exact likelihood evaluation and efficient gradient propagation that is useful during training. The recourse generator is trained with the objective of maximizing the likelihood under the accepted-class distribution while minimizing the likelihood under the denied-class distribution and other losses that encode recourse constraints. Furthermore, PAR includes a neighborhood-based conditioning mechanism to promote recourse generation that is customized to a factual. We validate PAR on widely used algorithmic recourse datasets and demonstrate its efficiency in generating recourses that are valid, similar to the factual, sparse, and highly plausible, yielding superior performance over existing state-of-the-art approaches.