Look at the Variance! Efficient Black-box Explanations with Sobol-based Sensitivity Analysis - Archive ouverte HAL Accéder directement au contenu
Communication Dans Un Congrès Année : 2021

Look at the Variance! Efficient Black-box Explanations with Sobol-based Sensitivity Analysis

Matthieu Cord

Résumé

We describe a novel attribution method which is grounded in Sensitivity Analysis and uses Sobol indices. Beyond modeling the individual contributions of image regions, Sobol indices provide an efficient way to capture higher-order interactions between image regions and their contributions to a neural network's prediction through the lens of variance. We describe an approach that makes the computation of these indices efficient for high-dimensional problems by using perturbation masks coupled with efficient estimators to handle the high dimensionality of images. Importantly, we show that the proposed method leads to favorable scores on standard benchmarks for vision (and language models) while drastically reducing the computing time compared to other black-box methods-even surpassing the accuracy of state-of-the-art white-box methods which require access to internal representations. Our code is freely available: github.com/fel-thomas/ Sobol-Attribution-Method.
Fichier non déposé

Dates et versions

hal-03997895 , version 1 (20-02-2023)

Identifiants

  • HAL Id : hal-03997895 , version 1

Citer

Matthieu Cord. Look at the Variance! Efficient Black-box Explanations with Sobol-based Sensitivity Analysis. NeurIPS, Dec 2021, Visio conference, France. ⟨hal-03997895⟩
11 Consultations
0 Téléchargements

Partager

Gmail Facebook X LinkedIn More