Inverse Classification for Comparison-based Interpretability in Machine Learning
This addresses interpretability in machine learning for users needing explanations without access to model or data details, but it is incremental as it builds on existing comparison-based approaches.
The paper tackles the problem of explaining classifier predictions when no information about the classifier or data is available, by proposing an instance-based method that finds minimal changes to alter predictions, with experimental results on two datasets demonstrating its relevance.
In the context of post-hoc interpretability, this paper addresses the task of explaining the prediction of a classifier, considering the case where no information is available, neither on the classifier itself, nor on the processed data (neither the training nor the test data). It proposes an instance-based approach whose principle consists in determining the minimal changes needed to alter a prediction: given a data point whose classification must be explained, the proposed method consists in identifying a close neighbour classified differently, where the closeness definition integrates a sparsity constraint. This principle is implemented using observation generation in the Growing Spheres algorithm. Experimental results on two datasets illustrate the relevance of the proposed approach that can be used to gain knowledge about the classifier.