Skip to Main content Skip to Navigation
Conference papers

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

Abstract : 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.
Document type :
Conference papers
Complete list of metadata

https://hal.archives-ouvertes.fr/hal-03473083
Contributor : Thomas FEL Connect in order to contact the contributor
Submitted on : Thursday, December 9, 2021 - 4:50:30 PM
Last modification on : Friday, August 5, 2022 - 3:33:43 PM

File

2020_12_XAI_Sobol__NeurIPS_202...
Files produced by the author(s)

Identifiers

  • HAL Id : hal-03473083, version 1

Citation

Thomas Fel, Rémi Cadène, Mathieu Chalvidal, Matthieu Cord, David Vigouroux, et al.. Look at the Variance! Efficient Black-box Explanations with Sobol-based Sensitivity Analysis. Conference on Neural Information Processing Systems (NeurIPS), Dec 2022, Sydney, Australia. ⟨hal-03473083⟩

Share

Metrics

Record views

61

Files downloads

5