Investigating the Corruption Robustness of Image Classifiers with Random Lp-norm Corruptions

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Siedel, Georg, Shao, Weijia, Vock, Silvia, Morozov, Andrey
Format: Preprint
Published: 2023
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866929358653882368
author Siedel, Georg
Shao, Weijia
Vock, Silvia
Morozov, Andrey
author_facet Siedel, Georg
Shao, Weijia
Vock, Silvia
Morozov, Andrey
contents Robustness is a fundamental property of machine learning classifiers required to achieve safety and reliability. In the field of adversarial robustness of image classifiers, robustness is commonly defined as the stability of a model to all input changes within a p-norm distance. However, in the field of random corruption robustness, variations observed in the real world are used, while p-norm corruptions are rarely considered. This study investigates the use of random p-norm corruptions to augment the training and test data of image classifiers. We evaluate the model robustness against imperceptible random p-norm corruptions and propose a novel robustness metric. We empirically investigate whether robustness transfers across different p-norms and derive conclusions on which p-norm corruptions a model should be trained and evaluated. We find that training data augmentation with a combination of p-norm corruptions significantly improves corruption robustness, even on top of state-of-the-art data augmentation schemes.
format Preprint
id arxiv_https___arxiv_org_abs_2305_05400
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle Investigating the Corruption Robustness of Image Classifiers with Random Lp-norm Corruptions
Siedel, Georg
Shao, Weijia
Vock, Silvia
Morozov, Andrey
Machine Learning
Artificial Intelligence
Computer Vision and Pattern Recognition
Robustness is a fundamental property of machine learning classifiers required to achieve safety and reliability. In the field of adversarial robustness of image classifiers, robustness is commonly defined as the stability of a model to all input changes within a p-norm distance. However, in the field of random corruption robustness, variations observed in the real world are used, while p-norm corruptions are rarely considered. This study investigates the use of random p-norm corruptions to augment the training and test data of image classifiers. We evaluate the model robustness against imperceptible random p-norm corruptions and propose a novel robustness metric. We empirically investigate whether robustness transfers across different p-norms and derive conclusions on which p-norm corruptions a model should be trained and evaluated. We find that training data augmentation with a combination of p-norm corruptions significantly improves corruption robustness, even on top of state-of-the-art data augmentation schemes.
title Investigating the Corruption Robustness of Image Classifiers with Random Lp-norm Corruptions
topic Machine Learning
Artificial Intelligence
Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2305.05400