Enhancing Adversarial Robustness via Score-Based Optimization
This work addresses adversarial robustness for AI safety, offering an incremental improvement over prior diffusion-based defenses by enhancing efficiency and performance.
The paper tackled the problem of adversarial attacks on deep neural networks by introducing ScoreOpt, a test-time optimization method using score-based priors, which achieved improved robustness and faster inference compared to existing defenses on datasets like CIFAR10, CIFAR100, and ImageNet.
Adversarial attacks have the potential to mislead deep neural network classifiers by introducing slight perturbations. Developing algorithms that can mitigate the effects of these attacks is crucial for ensuring the safe use of artificial intelligence. Recent studies have suggested that score-based diffusion models are effective in adversarial defenses. However, existing diffusion-based defenses rely on the sequential simulation of the reversed stochastic differential equations of diffusion models, which are computationally inefficient and yield suboptimal results. In this paper, we introduce a novel adversarial defense scheme named ScoreOpt, which optimizes adversarial samples at test-time, towards original clean data in the direction guided by score-based priors. We conduct comprehensive experiments on multiple datasets, including CIFAR10, CIFAR100 and ImageNet. Our experimental results demonstrate that our approach outperforms existing adversarial defenses in terms of both robustness performance and inference speed.