Summary of The Limits Of Differential Privacy in Online Learning, by Bo Li et al.
The Limits of Differential Privacy in Online Learning
by Bo Li, Wei Wang, Peng Ye
First submitted to arxiv on: 8 Nov 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: None
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary The research paper investigates the fundamental limits of differential privacy in online learning algorithms, presenting evidence that separates three types of constraints: no DP, pure DP, and approximate DP. The authors describe a hypothesis class that is online learnable under approximate DP but not online learnable under pure DP under adaptive adversarial setting, indicating the need for approximate DP when dealing with adaptive adversaries. Additionally, they prove that any private online learner must make an infinite number of mistakes for almost all hypothesis classes, generalizing previous results and showing a strong separation between private and non-private settings. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary The paper explores how to keep data private while still being able to learn from it. It looks at different ways to achieve this balance between privacy and usefulness, called differential privacy. The authors show that some methods are better than others for certain types of data and adversaries. They also prove that any method trying to be both private and accurate will make many mistakes. |
Keywords
» Artificial intelligence » Online learning
