How robust accuracy suffers from certified training with convex relaxations

• URL: http://arxiv.org/abs/2306.06995v1
• Date: Mon, 12 Jun 2023 09:45:21 GMT
• Title: How robust accuracy suffers from certified training with convex relaxations
• Authors: Piersilvio De Bartolomeis, Jacob Clarysse, Amartya Sanyal, Fanny Yang
• Abstract summary: Adrial attacks pose significant threats to deploying state-of-the-art classifiers in safety-critical applications. Two classes of methods have emerged to address this issue: empirical defences and certified defences. We systematically compare the standard and robust error of these two robust training paradigms across multiple computer vision tasks.
• Score: 12.292092677396347
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Adversarial attacks pose significant threats to deploying state-of-the-art classifiers in safety-critical applications. Two classes of methods have emerged to address this issue: empirical defences and certified defences. Although certified defences come with robustness guarantees, empirical defences such as adversarial training enjoy much higher popularity among practitioners. In this paper, we systematically compare the standard and robust error of these two robust training paradigms across multiple computer vision tasks. We show that in most tasks and for both $\mathscr{l}_\infty$-ball and $\mathscr{l}_2$-ball threat models, certified training with convex relaxations suffers from worse standard and robust error than adversarial training. We further explore how the error gap between certified and adversarial training depends on the threat model and the data distribution. In particular, besides the perturbation budget, we identify as important factors the shape of the perturbation set and the implicit margin of the data distribution. We support our arguments with extensive ablations on both synthetic and image datasets.

Related papers

• Certified Causal Defense with Generalizable Robustness [14.238441767523602]
  We propose a novel certified defense framework GLEAN, which incorporates a causal perspective into the generalization problem in certified defense. Our framework integrates a certifiable causal factor learning component to disentangle the causal relations and spurious correlations between input and label. On top of that, we design a causally certified defense strategy to handle adversarial attacks on latent causal factors.
  arXiv  Detail & Related papers  (2024-08-28T00:14:09Z)
• FullCert: Deterministic End-to-End Certification for Training and Inference of Neural Networks [62.897993591443594]
  FullCert is the first end-to-end certifier with sound, deterministic bounds. We experimentally demonstrate FullCert's feasibility on two datasets.
  arXiv  Detail & Related papers  (2024-06-17T13:23:52Z)
• Efficient Adversarial Training in LLMs with Continuous Attacks [99.5882845458567]
  Large language models (LLMs) are vulnerable to adversarial attacks that can bypass their safety guardrails. We propose a fast adversarial training algorithm (C-AdvUL) composed of two losses. C-AdvIPO is an adversarial variant of IPO that does not require utility data for adversarially robust alignment.
  arXiv  Detail & Related papers  (2024-05-24T14:20:09Z)
• Doubly Robust Instance-Reweighted Adversarial Training [107.40683655362285]
  We propose a novel doubly-robust instance reweighted adversarial framework. Our importance weights are obtained by optimizing the KL-divergence regularized loss function. Our proposed approach outperforms related state-of-the-art baseline methods in terms of average robust performance.
  arXiv  Detail & Related papers  (2023-08-01T06:16:18Z)
• On Evaluating the Adversarial Robustness of Semantic Segmentation Models [0.0]
  A number of adversarial training approaches have been proposed as a defense against adversarial perturbation. We show for the first time that a number of models in previous work that are claimed to be robust are in fact not robust at all. We then evaluate simple adversarial training algorithms that produce reasonably robust models even under our set of strong attacks.
  arXiv  Detail & Related papers  (2023-06-25T11:45:08Z)
• Raising the Bar for Certified Adversarial Robustness with Diffusion Models [9.684141378657522]
  In this work, we demonstrate that a similar approach can substantially improve deterministic certified defenses. One of our main insights is that the difference between the training and test accuracy of the original model, is a good predictor of the magnitude of the improvement. Our approach achieves state-of-the-art deterministic robustness certificates on CIFAR-10 for the $ell$ ($epsilon = 36/255$) and $ell_infty$ ($epsilon = 8/255$) threat models.
  arXiv  Detail & Related papers  (2023-05-17T17:29:10Z)
• Adaptive Feature Alignment for Adversarial Training [56.17654691470554]
  CNNs are typically vulnerable to adversarial attacks, which pose a threat to security-sensitive applications. We propose the adaptive feature alignment (AFA) to generate features of arbitrary attacking strengths. Our method is trained to automatically align features of arbitrary attacking strength.
  arXiv  Detail & Related papers  (2021-05-31T17:01:05Z)
• Adversarial Robustness under Long-Tailed Distribution [93.50792075460336]
  Adversarial robustness has attracted extensive studies recently by revealing the vulnerability and intrinsic characteristics of deep networks. In this work we investigate the adversarial vulnerability as well as defense under long-tailed distributions. We propose a clean yet effective framework, RoBal, which consists of two dedicated modules, a scale-invariant and data re-balancing.
  arXiv  Detail & Related papers  (2021-04-06T17:53:08Z)
• Certifiably-Robust Federated Adversarial Learning via Randomized Smoothing [16.528628447356496]
  In this paper, we incorporate smoothing techniques into federated adversarial training to enable data-private distributed learning. Our experiments show that such an advanced federated adversarial learning framework can deliver models as robust as those trained by the centralized training.
  arXiv  Detail & Related papers  (2021-03-30T02:19:45Z)
• Self-Progressing Robust Training [146.8337017922058]
  Current robust training methods such as adversarial training explicitly uses an "attack" to generate adversarial examples. We propose a new framework called SPROUT, self-progressing robust training. Our results shed new light on scalable, effective and attack-independent robust training methods.
  arXiv  Detail & Related papers  (2020-12-22T00:45:24Z)
• Asymptotic Behavior of Adversarial Training in Binary Classification [41.7567932118769]
  Adversarial training is considered to be the state-of-the-art method for defense against adversarial attacks. Despite being successful in practice, several problems in understanding performance of adversarial training remain open. We derive precise theoretical predictions for the minimization of adversarial training in binary classification.
  arXiv  Detail & Related papers  (2020-10-26T01:44:20Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.