Defense Against Adversarial Attacks Using Feature Scattering-based Adversarial Training

2019

247 citations recorded by the source corpus.

We introduce a feature scattering-based adversarial training approach for improving model robustness against adversarial attacks. Conventional adversarial training approaches leverage a supervised scheme (either targeted or non-targeted) in generating attacks for training, which typically suffer from issues such as label leaking as noted in recent works. Differently, the proposed approach generates adversarial images for training through feature scattering in the latent space, which is unsupervised in nature and avoids label leaking. More importantly, this new approach generates perturbed images in a collaborative fashion, taking the inter-sample relationships into consideration. We conduct analysis on model robustness and demonstrate the effectiveness of the proposed approach through extensively experiments on different datasets compared with state-of-the-art approaches.

Subfields: Artificial Intelligence, Electrical and Electronic Engineering, Signal Processing

Citations in the corpus

• Intriguing properties of adversarial training (2019, background): “…then proposed to further improve the model robustness (Xie et al., 2019b; Zhang et al., 2019b; Hendrycks et al., 2019a; Qin et al., 2019; Zhang & Wang, 2019; Carmon et al., 2019; Hendrycks et al., 2019b; Alayrac et al., 2019; Zhai et al., 2019) or accelerate the adversarial training process…”
• A Self-supervised Approach for Adversarial Robustness (2020, methodology): “Appendix C. Comparison with [19] Defense comparison: Feature Scattering (FS) [19] based AT remains model and task-specific.”
• Cross-Entropy Loss Functions: Theoretical Analysis and Applications (2023, methodology): “…et al., 2020), weight averaging/perturbation (Gowal et al., 2020; Wu et al., 2020; Tsai et al., 2021; Yu et al., 2018; Prabhu et al., 2019) and other techniques (Zhang & Wang, 2019; Qin et al., 2019; Goldblum et al., 2020; Song et al., 2019; Pang et al., 2020b; Lee et al., 2020; Qian et al., 2021).”
• Uncovering the Limits of Adversarial Training against Norm-Bounded Adversarial Examples (2020, background): “…that many of the defense strategies proposed in the literature (Papernot et al., 2015; Lu et al., 2017; Kannan et al., 2018; Tao et al., 2018; Zhang & Wang, 2019) were broken by stronger adversaries (Carlini & Wagner, 2016; 2017b; Athalye & Sutskever, 2017; Engstrom et al., 2018; Carlini,…”
• Boosting Adversarial Training with Hypersphere Embedding (2020, methodology): “Zhang & Wang (2019) apply a feature-scatter solver for the inner maximization problem of AT, which is different from PGD.”
• Boosting Decision-Based Black-Box Adversarial Attacks with Random Sign Flip (2020, methodology): “Attack performance against TRADES [53], FeatScatter [52], KWTA [47] and FeatDenoise [49].”