Adversarially Robust Training through Structured Gradient Regularization
arXiv:1805.08736
Abstract
We propose a novel data-dependent structured gradient regularizer to increase the robustness of neural networks vis-a-vis adversarial perturbations. Our regularizer can be derived as a controlled approximation from first principles, leveraging the fundamental link between training with noise and regularization. It adds very little computational overhead during learning and is simple to implement generically in standard deep learning frameworks. Our experiments provide strong evidence that structured gradient regularization can act as an effective first line of defense against attacks based on low-level signal corruption.
References in corpus (14)
- Adam: A Method for Stochastic Optimization
- Very Deep Convolutional Networks for Large-Scale Image Recognition
- Towards Deep Learning Models Resistant to Adversarial Attacks
- Transferability in Machine Learning: from Phenomena to Black-Box Attacks using Adversarial Samples
- Understanding deep learning requires rethinking generalization
- Delving into Transferable Adversarial Examples and Black-box Attacks
- The Space of Transferable Adversarial Examples
- Technical Report on the CleverHans v2.1.0 Adversarial Examples Library
- Distributional Smoothing with Virtual Adversarial Training
- Certifying Some Distributional Robustness with Principled Adversarial Training
- Virtual Adversarial Training: A Regularization Method for Supervised and Semi-Supervised Learning
- Stabilizing Training of Generative Adversarial Networks through Regularization
- Parseval Networks: Improving Robustness to Adversarial Examples
- Adversarial Training Versus Weight Decay
Cited by in corpus (8)
- Motivating the Rules of the Game for Adversarial Example Research
- Diversity can be Transferred: Output Diversification for White- and Black-box Attacks
- Adversarial Examples - A Complete Characterisation of the Phenomenon
- Adversarial Examples on Object Recognition: A Comprehensive Survey
- The Geometry of Adversarial Training in Binary Classification
- Proximal Mapping for Deep Regularization
- Reparameterized Variational Divergence Minimization for Stable Imitation
- Identifying and Exploiting Structures for Reliable Deep Learning