Probabilistic Margins for Instance Reweighting in Adversarial Training
arXiv:2106.07904
Abstract
Reweighting adversarial data during training has been recently shown to improve adversarial robustness, where data closer to the current decision boundaries are regarded as more critical and given larger weights. However, existing methods measuring the closeness are not very reliable: they are discrete and can take only a few values, and they are path-dependent, i.e., they may change given the same start and end points with different attack paths. In this paper, we propose three types of probabilistic margin (PM), which are continuous and path-independent, for measuring the aforementioned closeness and reweighting adversarial data. Specifically, a PM is defined as the difference between two estimated class-posterior probabilities, e.g., such the probability of the true label minus the probability of the most confusing label given some natural data. Though different PMs capture different geometric properties, all three PMs share a negative correlation with the vulnerability of data: data with larger/smaller PMs are safer/riskier and should have smaller/larger weights. Experiments demonstrate that PMs are reliable measurements and PM-based reweighting methods outperform state-of-the-art methods.
17 pages, 4 figures
References in corpus (11)
- Theoretically Principled Trade-off between Robustness and Accuracy
- On Evaluating Adversarial Robustness
- Fast is better than free: Revisiting adversarial training
- Imperceptible, Robust, and Targeted Adversarial Examples for Automatic Speech Recognition
- NATTACK: Learning the Distributions of Adversarial Examples for an Improved Black-Box Attack on Deep Neural Networks
- Adversarial Robustness: From Self-Supervised Pre-Training to Fine-Tuning
- Once-for-All Adversarial Training: In-Situ Tradeoff between Robustness and Accuracy for Free
- Maximum Mean Discrepancy Test is Aware of Adversarial Attacks
- Bag of Tricks for Adversarial Training
- PCAL: A Privacy-preserving Intelligent Credit Risk Modeling Framework Based on Adversarial Learning
- Projection & Probability-Driven Black-Box Attack