Private and Online Learnability Are Equivalent
Journal of the ACM(2022)
摘要
Let H be a binary-labeled concept class. We prove that H can be PAC learned by an (approximate) differentially private algorithm if and only if it has a finite Littlestone dimension. This implies a qualitative equivalence between online learnability and private PAC learnability.
更多查看译文
关键词
Differential privacy, PAC learning, online learning, Littlestone dimension
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