Generating Likely Counterfactuals Using Sum-Product Networks
This addresses the need for explainable AI under regulatory and user demands, offering an incremental improvement by balancing plausibility with traditional criteria like distance and sparsity.
The paper tackles the problem of generating plausible counterfactual explanations for AI decisions by proposing a system that produces high-likelihood, close, and sparse explanations using Mixed-Integer Optimization (MIO) with a Sum-Product Network (SPN) for likelihood estimation.
The need to explain decisions made by AI systems is driven by both recent regulation and user demand. The decisions are often explainable only post hoc. In counterfactual explanations, one may ask what constitutes the best counterfactual explanation. Clearly, multiple criteria must be taken into account, although "distance from the sample" is a key criterion. Recent methods that consider the plausibility of a counterfactual seem to sacrifice this original objective. Here, we present a system that provides high-likelihood explanations that are, at the same time, close and sparse. We show that the search for the most likely explanations satisfying many common desiderata for counterfactual explanations can be modeled using Mixed-Integer Optimization (MIO). We use a Sum-Product Network (SPN) to estimate the likelihood of a counterfactual. To achieve that, we propose an MIO formulation of an SPN, which can be of independent interest. The source code with examples is available at https://github.com/Epanemu/LiCE.