ADSAttack: An Adversarial Attack Algorithm via Searching Adversarial Distribution in Latent Space

Deep neural networks are susceptible to interference from deliberately crafted noise, which can lead incorrect classification results. Existing approaches make less use of latent space information and conduct pixel-domain modification in the input instead, increases computational cost decreases transferability. In this work, we propose an effective adversarial distribution searching-driven attack (ADSAttack) algorithm generate examples against deep networks. ADSAttack introduces affiliated network search for potential distributions image synthesizing examples. uses edge-detection locate low-level feature mapping sketch minimum disturbed area. Experimental results demonstrate that achieves higher transferability, better imperceptible visualization, faster generation speed compared traditional algorithms. To 1000 examples, takes 11.08s and, on average, a success rate 98.01%.