Towards Universal Adversarial Examples and Defenses
2021 IEEE Information Theory Workshop (ITW)(2021)
摘要
Adversarial examples have recently exposed the severe vulnerability of neural network models. However, most of the existing attacks require some form of target model information (i.e., weights/model inquiry/architecture) to improve the efficacy of the attack. We leverage the information-theoretic connections between robust learning and generalized rate-distortion theory to formulate a universal ad...
更多查看译文
关键词
Training,Costs,Computational modeling,Conferences,Neural networks,Rate-distortion,Inference algorithms
AI 理解论文
溯源树
样例
生成溯源树,研究论文发展脉络
Chat Paper
正在生成论文摘要