Saved in:
Bibliographic Details
Main Authors: Simpson, Lachlan, Millar, Kyle, Cheng, Adriel, Lim, Cheng-Chew, Chew, Hong Gunn
Format: Preprint
Published: 2024
Subjects:
Online Access:https://arxiv.org/abs/2402.18863
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866917608128774144
author Simpson, Lachlan
Millar, Kyle
Cheng, Adriel
Lim, Cheng-Chew
Chew, Hong Gunn
author_facet Simpson, Lachlan
Millar, Kyle
Cheng, Adriel
Lim, Cheng-Chew
Chew, Hong Gunn
contents Explainability models are now prevalent within machine learning to address the black-box nature of neural networks. The question now is which explainability model is most effective. Probabilistic Lipschitzness has demonstrated that the smoothness of a neural network is fundamentally linked to the quality of post hoc explanations. In this work, we prove theoretical lower bounds on the probabilistic Lipschitzness of Integrated Gradients, LIME and SmoothGrad. We propose a novel metric using probabilistic Lipschitzness, normalised astuteness, to compare the robustness of explainability models. Further, we prove a link between the local Lipschitz constant of a neural network and its stable rank. We then demonstrate that the stable rank of a neural network provides a heuristic for the robustness of explainability models.
format Preprint
id arxiv_https___arxiv_org_abs_2402_18863
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Probabilistic Lipschitzness and the Stable Rank for Comparing Explanation Models
Simpson, Lachlan
Millar, Kyle
Cheng, Adriel
Lim, Cheng-Chew
Chew, Hong Gunn
Machine Learning
Explainability models are now prevalent within machine learning to address the black-box nature of neural networks. The question now is which explainability model is most effective. Probabilistic Lipschitzness has demonstrated that the smoothness of a neural network is fundamentally linked to the quality of post hoc explanations. In this work, we prove theoretical lower bounds on the probabilistic Lipschitzness of Integrated Gradients, LIME and SmoothGrad. We propose a novel metric using probabilistic Lipschitzness, normalised astuteness, to compare the robustness of explainability models. Further, we prove a link between the local Lipschitz constant of a neural network and its stable rank. We then demonstrate that the stable rank of a neural network provides a heuristic for the robustness of explainability models.
title Probabilistic Lipschitzness and the Stable Rank for Comparing Explanation Models
topic Machine Learning
url https://arxiv.org/abs/2402.18863