Improving Model Robustness with Latent Distribution Locally and Globally
arXiv:2107.04401
Abstract
In this work, we consider model robustness of deep neural networks against adversarial attacks from a global manifold perspective. Leveraging both the local and global latent information, we propose a novel adversarial training method through robust optimization, and a tractable way to generate Latent Manifold Adversarial Examples (LMAEs) via an adversarial game between a discriminator and a classifier. The proposed adversarial training with latent distribution (ATLD) method defends against adversarial attacks by crafting LMAEs with the latent manifold in an unsupervised manner. ATLD preserves the local and global information of latent manifold and promises improved robustness against adversarial attacks. To verify the effectiveness of our proposed method, we conduct extensive experiments over different datasets (e.g., CIFAR-10, CIFAR-100, SVHN) with different adversarial attacks (e.g., PGD, CW), and show that our method substantially outperforms the state-of-the-art (e.g., Feature Scattering) in adversarial robustness by a large accuracy margin. The source codes are available at https://github.com/LitterQ/ATLD-pytorch.
References in corpus (23)
- Explaining and Harnessing Adversarial Examples
- Conditional Image Synthesis With Auxiliary Classifier GANs
- Wide Residual Networks
- Towards Deep Learning Models Resistant to Adversarial Attacks
- Conditional Adversarial Domain Adaptation
- Theoretically Principled Trade-off between Robustness and Accuracy
- Estimating divergence functionals and the likelihood ratio by convex risk minimization
- Reliable evaluation of adversarial robustness with an ensemble of diverse parameter-free attacks
- Adversarially Robust Generalization Requires More Data
- Using Pre-Training Can Improve Model Robustness and Uncertainty
- On the (Statistical) Detection of Adversarial Examples
- PixelDefend: Leveraging Generative Models to Understand and Defend against Adversarial Examples
- Mode Regularized Generative Adversarial Networks
- Virtual Adversarial Training: A Regularization Method for Supervised and Semi-Supervised Learning
- Uncovering the Limits of Adversarial Training against Norm-Bounded Adversarial Examples
- Image Synthesis with a Single (Robust) Classifier
- Adversarially Robust Generalization Just Requires More Unlabeled Data
- Metric Learning for Adversarial Robustness
- Adversarial Examples for Semantic Image Segmentation
- Overfitting in adversarially robust deep learning
- Do Wider Neural Networks Really Help Adversarial Robustness?
- Harnessing the Vulnerability of Latent Layers in Adversarially Trained Models
- Training Robust Deep Neural Networks via Adversarial Noise Propagation