A Modified Perturbed Sampling Method for Local Interpretable Model-agnostic Explanation
This work addresses interpretability in AI for users needing explanations of black-box models, but it is incremental as it modifies an existing method.
The paper tackled the problem of defective sampling in Local Interpretable Model-agnostic Explanation (LIME) by proposing a Modified Perturbed Sampling method (MPS-LIME) that addresses feature correlation issues, resulting in improved performance in understandability, fidelity, and efficiency for image classification.
Explainability is a gateway between Artificial Intelligence and society as the current popular deep learning models are generally weak in explaining the reasoning process and prediction results. Local Interpretable Model-agnostic Explanation (LIME) is a recent technique that explains the predictions of any classifier faithfully by learning an interpretable model locally around the prediction. However, the sampling operation in the standard implementation of LIME is defective. Perturbed samples are generated from a uniform distribution, ignoring the complicated correlation between features. This paper proposes a novel Modified Perturbed Sampling operation for LIME (MPS-LIME), which is formalized as the clique set construction problem. In image classification, MPS-LIME converts the superpixel image into an undirected graph. Various experiments show that the MPS-LIME explanation of the black-box model achieves much better performance in terms of understandability, fidelity, and efficiency.