paper-with-me

홈 › Papers

Adversarial Training Makes Weight Loss Landscape Sharper in Logistic Regression

2021-02-05 · Masanori Yamada, Sekitoshi Kanai, Tomoharu Iwata, Tomokatsu Takahashi, Yuki Yamanaka, Hiroshi Takahashi, Atsutoshi Kumagai

Adversarial training is actively studied for learning robust models against adversarial examples. A recent study finds that adversarially trained models degenerate generalization performance on adversarial examples when their weight loss landscape, which is loss changes with respect to weights, is sharp. Unfortunately, it has been experimentally shown that adversarial training sharpens the weight loss landscape, but this phenomenon has not been theoretically clarified. Therefore, we theoretically analyze this phenomenon in this paper. As a first step, this paper proves that adversarial training with the L2 norm constraints sharpens the weight loss landscape in the linear logistic regression model. Our analysis reveals that the sharpness of the weight loss landscape is caused by the noise aligned in the direction of increasing the loss, which is used in adversarial training. We theoretically and experimentally confirm that the weight loss landscape becomes sharper as the magnitude of the noise of adversarial training increases in the linear logistic regression model. Moreover, we experimentally confirm the same phenomena in ResNet18 with softmax as a more general case.

📄 PDF Abstract BibTeX arXiv:2102.02950

Code (0)

등록된 구현이 없습니다.

Tasks

regression

Methods 이 논문이 사용한 방법론

Softmax The Softmax output function transforms a previous layer's output into a vector of probabilities. It is commonly used for multiclass classification. Given an input vector $x$…
Logistic Regression Logistic Regression, despite its name, is a linear model for classification rather than regression. Logistic regression is also known in the literature as logit regression,…

Similar Papers 제목 키워드 기반

Adversarial Weight Perturbation Helps Robust Generalization

2020-04-13 · NeurIPS 2020 12 · Dongxian Wu, Shu-Tao Xia, Yisen Wang

The study on improving the robustness of deep neural networks against adversarial examples grows rapidly in recent years. Among them, adversarial training is the most promising one, which flattens the input loss landscap…

Adversarial Robustness

Understanding and Combating Robust Overfitting via Input Loss Landscape Analysis and Regularization

2022-12-09 · Lin Li, Michael Spratling

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 o…

Adversarial Attack

CAP: Co-Adversarial Perturbation on Weights and Features for Improving Generalization of Graph Neural Networks

2021-10-28 · Haotian Xue, Kaixiong Zhou, Tianlong Chen, Kai Guo 외

Despite the recent advances of graph neural networks (GNNs) in modeling graph data, the training of GNNs on large datasets is notoriously hard due to the overfitting. Adversarial training, which augments data with the wo…

Relating Adversarially Robust Generalization to Flat Minima

2021-04-09 · ICCV 2021 10 · David Stutz, Matthias Hein, Bernt Schiele

Adversarial training (AT) has become the de-facto standard to obtain models robust against adversarial examples. However, AT exhibits severe robust overfitting: cross-entropy loss on adversarial examples, so-called robus…

Adversarial Robustness

AdvRush: Searching for Adversarially Robust Neural Architectures

2021-08-03 · ICCV 2021 10 · Jisoo Mok, Byunggook Na, Hyeokjun Choe, Sungroh Yoon

Deep neural networks continue to awe the world with their remarkable performance. Their predictions, however, are prone to be corrupted by adversarial examples that are imperceptible to humans. Current efforts to improve…

Adversarial RobustnessNeural Architecture Search