Towards Efficient Data-Centric Robust Machine Learning with Noise-based Augmentation
Abstract
The data-centric machine learning aims to find effective ways to build appropriate datasets which can improve the performance of AI models. In this paper, we mainly focus on designing an efficient data-centric scheme to improve robustness for models towards unforeseen malicious inputs in the black-box test settings. Specifically, we introduce a noised-based data augmentation method which is composed of Gaussian Noise, Salt-and-Pepper noise, and the PGD adversarial perturbations. The proposed method is built on lightweight algorithms and proved highly effective based on comprehensive evaluations, showing good efficiency on computation cost and robustness enhancement. In addition, we share our insights about the data-centric robust machine learning gained from our experiments.
Cite
@article{arxiv.2203.03810,
title = {Towards Efficient Data-Centric Robust Machine Learning with Noise-based Augmentation},
author = {Xiaogeng Liu and Haoyu Wang and Yechao Zhang and Fangzhou Wu and Shengshan Hu},
journal= {arXiv preprint arXiv:2203.03810},
year = {2022}
}
Comments
Competition paper of AAAI2022:Data-Centric Robust Learning on ML Models. Published in Workshop on Adversarial Machine Learning and Beyond at AAAI2022