Enhancing Adversarial Defense by k-Winners-Take-All

Abstract

We propose a simple change to existing neural network structures for better defending against gradient-based adversarial attacks. Instead of using popular activation functions (such as ReLU), we advocate the use of k-Winners-Take-All (k-WTA) activation, a C0 discontinuous function that purposely invalidates the neural network model's gradient at densely distributed input data points. The proposed k-WTA activation can be readily used in nearly all existing networks and training methods with no significant overhead. Our proposal is theoretically rationalized. We analyze why the discontinuities in k-WTA networks can largely prevent gradient-based search of adversarial examples and why they at the same time remain innocuous to the network training. This understanding is also empirically backed. We test k-WTA activation on various network structures optimized by a training method, be it adversarial training or not. In all cases, the robustness of k-WTA networks outperforms that of traditional networks under white-box attacks.

References in corpus (18)

• Very Deep Convolutional Networks for Large-Scale Image Recognition
• Ensemble Adversarial Training: Attacks and Defenses
• Wide Residual Networks
• Theoretically Principled Trade-off between Robustness and Accuracy
• On the Number of Linear Regions of Deep Neural Networks
• A Convergence Theory for Deep Learning via Over-Parameterization
• Synthesizing Robust Adversarial Examples
• Characterizing Adversarial Subspaces Using Local Intrinsic Dimensionality
• Gradient Descent Provably Optimizes Over-parameterized Neural Networks
• Robustness May Be at Odds with Accuracy
• Improving the Adversarial Robustness and Interpretability of Deep Neural Networks by Regularizing their Input Gradients
• Learning Overparameterized Neural Networks via Stochastic Gradient Descent on Structured Data
• Simple Black-Box Adversarial Perturbations for Deep Networks
• On the Convergence Rate of Training Recurrent Neural Networks
• Feature Denoising for Improving Adversarial Robustness
• HopSkipJumpAttack: A Query-Efficient Decision-Based Attack
• Quadratic Suffices for Over-parametrization via Matrix Chernoff Bound
• Understanding Locally Competitive Networks

Cited by in corpus (4)

• RP2K: A Large-Scale Retail Product Dataset for Fine-Grained Image Classification
• Picket: Guarding Against Corrupted Data in Tabular Data during Learning and Inference
• Meta Gradient Adversarial Attack
• One Man's Trash is Another Man's Treasure: Resisting Adversarial Examples by Adversarial Examples