paper-with-me

홈 › Papers

Inducing Data Amplification Using Auxiliary Datasets in Adversarial Training

2022-09-27 · Saehyung Lee, Hyungyu Lee

Several recent studies have shown that the use of extra in-distribution data can lead to a high level of adversarial robustness. However, there is no guarantee that it will always be possible to obtain sufficient extra data for a selected dataset. In this paper, we propose a biased multi-domain adversarial training (BiaMAT) method that induces training data amplification on a primary dataset using publicly available auxiliary datasets, without requiring the class distribution match between the primary and auxiliary datasets. The proposed method can achieve increased adversarial robustness on a primary dataset by leveraging auxiliary datasets via multi-domain learning. Specifically, data amplification on both robust and non-robust features can be accomplished through the application of BiaMAT as demonstrated through a theoretical and empirical analysis. Moreover, we demonstrate that while existing methods are vulnerable to negative transfer due to the distributional discrepancy between auxiliary and primary data, the proposed method enables neural networks to flexibly leverage diverse image datasets for adversarial training by successfully handling the domain discrepancy through the application of a confidence-based selection strategy. The pre-trained models and code are available at: \url{https://github.com/Saehyung-Lee/BiaMAT}.

📄 PDF Abstract BibTeX arXiv:2209.14053

Code (1)

saehyung-lee/biamat 공식 구현 pytorch

Tasks

Adversarial Robustness

Similar Papers 제목 키워드 기반

Biased Multi-Domain Adversarial Training

2021-09-29 · Saehyung Lee, Hyungyu Lee, Sanghyuk Chun, Sungroh Yoon

Several recent studies have shown that the use of extra in-distribution data can lead to a high level of adversarial robustness. However, there is no guarantee that it will always be possible to obtain sufficient extra d…

Adversarial Robustness

Detecting Adversarial Data via Provable Adversarial Noise Amplification

2026-05-04 · Furkan Mumcu, Yasin Yilmaz arxiv

The nonuniform and growing impact of adversarial noise across the layers of deep neural networks has been used in the literature, without a formal mathematical justification, to detect adversarial inputs and improve robu…

Adversarial Defense

Improving Adversarial Robustness via Activation Amplification and Attenuation

2026-06-26 · Taïga Gonçalves, Yongsong Huang, Tomo Miyazaki, Shinichiro Omachi arxiv

The existence of adversarial attacks is often attributed to the presence of non-robust features in neural networks. While prior defenses reduce their impact via pruning, masking, or feature recalibration, we instead prop…

Adversarial Robustness

Mitigating attribute amplification in counterfactual image generation

2024-03-14 · Tian Xia, Mélanie Roschewitz, Fabio De Sousa Ribeiro, Charles Jones 외

Causal generative modelling is gaining interest in medical imaging due to its ability to answer interventional and counterfactual queries. Most work focuses on generating counterfactual images that look plausible, using …

AttributecounterfactualImage Generation

Making Bias Amplification in Balanced Datasets Directional and Interpretable

2024-12-15 · Bhanu Tokas, Rahul Nair, Hannah Kerner

Most of the ML datasets we use today are biased. When we train models on these biased datasets, they often not only learn dataset biases but can also amplify them -- a phenomenon known as bias amplification. Several co-o…

Attribute