English
Related papers

Related papers: Understanding Catastrophic Overfitting in Adversar…

200 papers

In this paper, we show that Generative Adversarial Networks (GANs) suffer from catastrophic forgetting even when they are trained to approximate a single target distribution. We show that GAN training is a continual learning problem in…

Machine Learning · Computer Science 2020-03-24 Hoang Thanh-Tung , Truyen Tran

Traditional classification algorithms assume that training and test data come from similar distributions. This assumption is violated in adversarial settings, where malicious actors modify instances to evade detection. A number of custom…

Computer Science and Game Theory · Computer Science 2016-11-29 Bo Li , Yevgeniy Vorobeychik , Xinyun Chen

Adversarial training suffers from robust overfitting, a phenomenon where the robust test accuracy starts to decrease during training. In this paper, we focus on reducing robust overfitting by using common data augmentation schemes. We…

Computer Vision and Pattern Recognition · Computer Science 2021-11-10 Sylvestre-Alvise Rebuffi , Sven Gowal , Dan A. Calian , Florian Stimberg , Olivia Wiles , Timothy Mann

Adversarial training (AT) constructs robust neural networks by incorporating adversarial perturbations into natural data. However, it is plagued by the issue of robust overfitting (RO), which severely damages the model's robustness. In this…

Machine Learning · Computer Science 2024-07-30 Chaojian Yu , Xiaolong Shi , Jun Yu , Bo Han , Tongliang Liu

Adversarial training is by far the most successful strategy for improving robustness of neural networks to adversarial attacks. Despite its success as a defense mechanism, adversarial training fails to generalize well to unperturbed test…

Machine Learning · Computer Science 2019-10-18 Yogesh Balaji , Tom Goldstein , Judy Hoffman

Supervised fine-tuning (SFT) is a common first stage of LLM post-training, teaching the model to follow instructions and shaping its behavior as a helpful assistant. At the same time, SFT may harm the fundamental capabilities of an LLM,…

Machine Learning · Computer Science 2026-04-16 Mark Rofin , Aditya Varre , Nicolas Flammarion

Contrastive Forward-Forward (CFF) learning trains Vision Transformers layer by layer against supervised contrastive objectives. CFF training can be sensitive to random seed, but the sources of this instability are poorly understood. We…

Machine Learning · Computer Science 2026-03-03 Joshua Steier

It is well known that adversarial attacks can fool deep neural networks with imperceptible perturbations. Although adversarial training significantly improves model robustness, failure cases of defense still broadly exist. In this work, we…

Machine Learning · Computer Science 2021-06-10 Boxi Wu , Heng Pan , Li Shen , Jindong Gu , Shuai Zhao , Zhifeng Li , Deng Cai , Xiaofei He , Wei Liu

Continual learning focuses on learning non-stationary data distribution without forgetting previous knowledge. Rehearsal-based approaches are commonly used to combat catastrophic forgetting. However, these approaches suffer from a problem…

Machine Learning · Computer Science 2024-05-21 Hikmat Khan , Ghulam Rasool , Nidhal Carla Bouaynaya

Defenses against adversarial examples, such as adversarial training, are typically tailored to a single perturbation type (e.g., small $\ell_\infty$-noise). For other perturbations, these defenses offer no guarantees and, at times, even…

Machine Learning · Computer Science 2019-10-21 Florian Tramèr , Dan Boneh

It has been consistently reported that many machine learning models are susceptible to adversarial attacks i.e., small additive adversarial perturbations applied to data points can cause misclassification. Adversarial training using…

Machine Learning · Statistics 2021-07-15 Hossein Taheri , Ramtin Pedarsani , Christos Thrampoulidis

Current SOTA adversarially robust models are mostly based on adversarial training (AT) and differ only by some regularizers either at inner maximization or outer minimization steps. Being repetitive in nature during the inner maximization…

Machine Learning · Computer Science 2021-11-02 Anindya Sarkar , Anirban Sarkar , Sowrya Gali , Vineeth N Balasubramanian

Several recent works have developed methods for training classifiers that are certifiably robust against norm-bounded adversarial perturbations. These methods assume that all the adversarial transformations are equally important, which is…

Machine Learning · Computer Science 2019-03-06 Xiao Zhang , David Evans

Adversarial training is widely used to improve the robustness of deep neural networks to adversarial attack. However, adversarial training is prone to overfitting, and the cause is far from clear. This work sheds light on the mechanisms…

Machine Learning · Computer Science 2022-12-12 Lin Li , Michael Spratling

Adversarial examples are a pervasive phenomenon of machine learning models where seemingly imperceptible perturbations to the input lead to misclassifications for otherwise statistically accurate models. We propose a geometric framework,…

Machine Learning · Computer Science 2019-05-06 Marc Khoury , Dylan Hadfield-Menell

Quantum classifiers are vulnerable to adversarial attacks that manipulate their input classical or quantum data. A promising countermeasure is adversarial training, where quantum classifiers are trained by using an attack-aware, adversarial…

Quantum Physics · Physics 2025-04-25 Petros Georgiou , Aaron Mark Thomas , Sharu Theresa Jose , Osvaldo Simeone

This paper revisits the simple, long-studied, yet still unsolved problem of making image classifiers robust to imperceptible perturbations. Taking CIFAR10 as an example, SOTA clean accuracy is about $100$%, but SOTA robustness to…

Machine Learning · Computer Science 2024-07-11 Brian R. Bartoldson , James Diffenderfer , Konstantinos Parasyris , Bhavya Kailkhura

Similar to surprising performance in the standard deep learning, deep nets trained by adversarial training also generalize well for unseen clean data (natural data). However, despite adversarial training can achieve low robust training…

Machine Learning · Computer Science 2025-05-23 Binghui Li , Yuanzhi Li

Pretrained models from self-supervision are prevalently used in fine-tuning downstream tasks faster or for better accuracy. However, gaining robustness from pretraining is left unexplored. We introduce adversarial training into…

Computer Vision and Pattern Recognition · Computer Science 2020-03-31 Tianlong Chen , Sijia Liu , Shiyu Chang , Yu Cheng , Lisa Amini , Zhangyang Wang

Transfer learning is often used to decrease the computational cost of model training, as fine-tuning a model allows a downstream task to leverage the features learned from the pre-training dataset and quickly adapt them to a new task. This…

Machine Learning · Computer Science 2025-07-01 Joshua C. Zhao , Saurabh Bagchi