A Two-Stage Algorithm for Cost-Efficient Multi-instance Counterfactual Explanations
This addresses the need for multi-instance explanations in real-world applications like customer satisfaction, though it appears to be an incremental improvement over existing single-instance methods.
The paper tackles the problem of generating counterfactual explanations that work for multiple instances simultaneously, proposing a two-stage algorithm that identifies groups of instances and computes cost-efficient multi-instance counterfactuals. The result is demonstrated through comparative evaluation against popular alternatives, though no specific performance numbers are provided in the abstract.
Counterfactual explanations constitute among the most popular methods for analyzing black-box systems since they can recommend cost-efficient and actionable changes to the input of a system to obtain the desired system output. While most of the existing counterfactual methods explain a single instance, several real-world problems, such as customer satisfaction, require the identification of a single counterfactual that can satisfy multiple instances (e.g. customers) simultaneously. To address this limitation, in this work, we propose a flexible two-stage algorithm for finding groups of instances and computing cost-efficient multi-instance counterfactual explanations. The paper presents the algorithm and its performance against popular alternatives through a comparative evaluation.