paper-with-me

Papers

Evaluating the Robustness of Interpretability Methods through Explanation Invariance and Equivariance

2023-04-13 · NeurIPS 2023 11 · Jonathan Crabbé, Mihaela van der Schaar

Interpretability methods are valuable only if their explanations faithfully describe the explained model. In this work, we consider neural networks whose predictions are invariant under a specific symmetry group. This includes popular architectures, ranging from convolutional to graph neural networks. Any explanation that faithfully explains this type of model needs to be in agreement with this invariance property. We formalize this intuition through the notion of explanation invariance and equivariance by leveraging the formalism from geometric deep learning. Through this rigorous formalism, we derive (1) two metrics to measure the robustness of any interpretability method with respect to the model symmetry group; (2) theoretical robustness guarantees for some popular interpretability methods and (3) a systematic approach to increase the invariance of any interpretability method with respect to a symmetry group. By empirically measuring our metrics for explanations of models associated with various modalities and symmetry groups, we derive a set of 5 guidelines to allow users and developers of interpretability methods to produce robust explanations.

📄 PDF Abstract BibTeX arXiv:2304.06715

Code (2)

jonathancrabbe/robustxai 공식 구현 pytorch
vanderschaarlab/robustxai 공식 구현 pytorch

Similar Papers 제목 키워드 기반

Secure Diagnostics: Adversarial Robustness Meets Clinical Interpretability

2025-04-07 · Mohammad Hossein Najafi, Mohammad Morsali, Mohammadreza Pashanejad, Saman Soleimani Roudi 외

Deep neural networks for medical image classification often fail to generalize consistently in clinical practice due to violations of the i.i.d. assumption and opaque decision-making. This paper examines interpretability…

Adversarial AttackAdversarial RobustnessDecision MakingFracture detection+3

xai_evals : A Framework for Evaluating Post-Hoc Local Explanation Methods

2025-02-05 · Pratinav Seth, Yashwardhan Rathore, Neeraj Kumar Singh, Chintan Chitroda 외

The growing complexity of machine learning and deep learning models has led to an increased reliance on opaque "black box" systems, making it difficult to understand the rationale behind predictions. This lack of transpa…

Benchmarking

Evaluating Explanations: An Explanatory Virtues Framework for Mechanistic Interpretability -- The Strange Science Part I.ii

2025-05-02 · Kola Ayonrinde, Louis Jaburi

Mechanistic Interpretability (MI) aims to understand neural networks through causal explanations. Though MI has many explanation-generating methods, progress has been limited by the lack of a universal approach to evalua…

Philosophy

On the Robustness of Interpretability Methods

2018-06-21 · David Alvarez-Melis, Tommi S. Jaakkola

We argue that robustness of explanations---i.e., that similar inputs should give rise to similar explanations---is a key desideratum for interpretability. We introduce metrics to quantify robustness and demonstrate that …

SAFARI: Versatile and Efficient Evaluations for Robustness of Interpretability

2022-08-19 · ICCV 2023 1 · Wei Huang, Xingyu Zhao, Gaojie Jin, Xiaowei Huang

Interpretability of Deep Learning (DL) is a barrier to trustworthy AI. Despite great efforts made by the Explainable AI (XAI) community, explanations lack robustness -- indistinguishable input perturbations may lead to d…

Explainable Artificial Intelligence (XAI)