摘要: Summary:《Adversarial Machine Learning in Image Classification: A Survey Towards the Defender’s Perspective》在前半部分对抗攻击方面的小结 Note “Taxonomy of Adversarial Images” (Machado 等, 2023, p. 5) (pdf) 扰动范围(Perturbation Scope): 个体扰动(Individual-scoped perturbations): 为每个输入图像单独生成的扰动。 通 阅读全文
posted @ 2024-06-11 20:18 Rosmontis 阅读(13) 评论(0) 推荐(0) 编辑