paper-with-me

홈 › Papers

AFD: Mitigating Feature Gap for Adversarial Robustness by Feature Disentanglement

2024-01-26 · Nuoyan Zhou, Dawei Zhou, Decheng Liu, Nannan Wang, Xinbo Gao

Adversarial fine-tuning methods enhance adversarial robustness via fine-tuning the pre-trained model in an adversarial training manner. However, we identify that some specific latent features of adversarial samples are confused by adversarial perturbation and lead to an unexpectedly increasing gap between features in the last hidden layer of natural and adversarial samples. To address this issue, we propose a disentanglement-based approach to explicitly model and further remove the specific latent features. We introduce a feature disentangler to separate out the specific latent features from the features of the adversarial samples, thereby boosting robustness by eliminating the specific latent features. Besides, we align clean features in the pre-trained model with features of adversarial samples in the fine-tuned model, to benefit from the intrinsic features of natural samples. Empirical evaluations on three benchmark datasets demonstrate that our approach surpasses existing adversarial fine-tuning methods and adversarial training baselines.

📄 PDF Abstract BibTeX arXiv:2401.14707

Code (1)

changzhang777/AFD 공식 구현 pytorch

Tasks

Adversarial RobustnessDisentanglement

Methods 이 논문이 사용한 방법론

ALIGN In the ALIGN method, visual and language representations are jointly trained from noisy image alt-text data. The image and text encoders are learned via contrastive loss…

Similar Papers 제목 키워드 기반

ERIS: An Energy-Guided Feature Disentanglement Framework for Out-of-Distribution Time Series Classification

2025-08-19 · Xin Wu, Fei Teng, Ji Zhang, Xingwang Li 외 arxiv

An ideal time series classification (TSC) should be able to capture invariant representations, but achieving reliable performance on out-of-distribution (OOD) data remains a core obstacle. This obstacle arises from the w…

Time Series Classification

Mitigating Shortcut Learning via Feature Disentanglement in Medical Imaging: A Benchmark Study

2026-02-17 · Sarah Müller, Philipp Berens arxiv

Although deep learning models in medical imaging often achieve excellent classification performance, they can rely on shortcut learning, exploiting spurious correlations or confounding factors that are not causally relat…

Computational Efficiency

Mitigating Low-Frequency Bias: Feature Recalibration and Frequency Attention Regularization for Adversarial Robustness

2024-07-04 · Kejia Zhang, Juanjuan Weng, Yuanzheng Cai, Zhiming Luo 외

Ensuring the robustness of deep neural networks against adversarial attacks remains a fundamental challenge in computer vision. While adversarial training (AT) has emerged as a promising defense strategy, our analysis re…

Adversarial RobustnessDisentanglement

Decoder-free Robustness Disentanglement without (Additional) Supervision

2020-07-02 · Yifei Wang, Dan Peng, Furui Liu, Zhenguo Li 외

Adversarial Training (AT) is proposed to alleviate the adversarial vulnerability of machine learning models by extracting only robust features from the input, which, however, inevitably leads to severe accuracy reduction…

BIG-bench Machine LearningDecoderDisentanglementRepresentation Learning

Exploring Robust Features for Improving Adversarial Robustness

2023-09-09 · Hong Wang, Yuefan Deng, Shinjae Yoo, Yuewei Lin

While deep neural networks (DNNs) have revolutionized many fields, their fragility to carefully designed adversarial attacks impedes the usage of DNNs in safety-critical applications. In this paper, we strive to explore …

Adversarial RobustnessDisentanglement