LGAIJun 2, 2022

Which Explanation Should I Choose? A Function Approximation Perspective to Characterizing Post Hoc Explanations

Harvard
arXiv:2206.01254v3116 citationsh-index: 43
Originality Incremental advance
AI Analysis

This work addresses the practical challenge of choosing explanation methods in machine learning interpretability, offering a unifying perspective that is incremental but clarifies existing methods.

The paper tackles the lack of a common foundational goal among post hoc explanation methods by unifying eight popular methods under a local function approximation framework, showing that no method performs optimally across all scenarios and providing a principled approach for selection based on faithfulness.

A critical problem in the field of post hoc explainability is the lack of a common foundational goal among methods. For example, some methods are motivated by function approximation, some by game theoretic notions, and some by obtaining clean visualizations. This fragmentation of goals causes not only an inconsistent conceptual understanding of explanations but also the practical challenge of not knowing which method to use when. In this work, we begin to address these challenges by unifying eight popular post hoc explanation methods (LIME, C-LIME, KernelSHAP, Occlusion, Vanilla Gradients, Gradients x Input, SmoothGrad, and Integrated Gradients). We show that these methods all perform local function approximation of the black-box model, differing only in the neighbourhood and loss function used to perform the approximation. This unification enables us to (1) state a no free lunch theorem for explanation methods, demonstrating that no method can perform optimally across all neighbourhoods, and (2) provide a guiding principle to choose among methods based on faithfulness to the black-box model. We empirically validate these theoretical results using various real-world datasets, model classes, and prediction tasks. By bringing diverse explanation methods into a common framework, this work (1) advances the conceptual understanding of these methods, revealing their shared local function approximation objective, properties, and relation to one another, and (2) guides the use of these methods in practice, providing a principled approach to choose among methods and paving the way for the creation of new ones.

Code Implementations1 repo
Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes