Adversarial attacks and robust defenses in deep learning
Deep learning models have shown exceptional performance in many applications, including
computer vision, natural language processing, and speech processing. However, if no
defense strategy is considered, deep learning models are vulnerable to adversarial attacks.
In this chapter, we will first describe various typical adversarial attacks. Then we will
describe different adversarial defense methods for image classification and object detection
tasks.
computer vision, natural language processing, and speech processing. However, if no
defense strategy is considered, deep learning models are vulnerable to adversarial attacks.
In this chapter, we will first describe various typical adversarial attacks. Then we will
describe different adversarial defense methods for image classification and object detection
tasks.
以上显示的是最相近的搜索结果。 查看全部搜索结果