Summary of Efficient Availability Attacks Against Supervised and Contrastive Learning Simultaneously, by Yihan Wang and Yifan Zhu and Xiao-shan Gao
Efficient Availability Attacks against Supervised and Contrastive Learning Simultaneously
by Yihan Wang, Yifan Zhu, Xiao-Shan Gao
First submitted to arxiv on: 6 Feb 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: Machine Learning (stat.ML)
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 paper presents a novel approach to availability attacks that generate imperceptible noise and make unlearnable examples to protect private data and commercial datasets. The proposed AUE and AAP attacks achieve state-of-the-art worst-case unlearnability across supervised learning (SL) and contrastive learning (CL) algorithms with less computation consumption, making them suitable for real-world applications. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This paper is about protecting sensitive data from being used by unauthorized people or algorithms. It shows that most existing methods aren’t good at keeping both types of data (supervised and contrastive) safe. The researchers came up with a new way to make examples unlearnable, which makes it harder for bad actors to get away with using the data. This is important because it helps keep our personal information private. |
Keywords
* Artificial intelligence * Supervised
