Summary of Hessian-free Online Certified Unlearning, by Xinbao Qiao et al.
Hessian-Free Online Certified Unlearning
by Xinbao Qiao, Meng Zhang, Ming Tang, Ermin Wei
First submitted to arxiv on: 2 Apr 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: Artificial Intelligence (cs.AI)
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 proposes an efficient approach to machine unlearning, enabling models to selectively forget specific data. The authors improve upon previous methods by pre-computing and storing statistics extracted from second-order information and implementing unlearning through Newton-style updates. The proposed method outperforms state-of-the-art methods in terms of unlearning and generalization guarantees, deletion capacity, and time/storage complexity. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary Machine learning models can be trained to forget specific data, allowing them to respect the “right to be forgotten” of data owners. This paper introduces a new approach to machine unlearning that is fast and efficient. The method uses statistical vectors to keep track of training data and can remove data in just milliseconds. The results show that this approach is much faster and more accurate than existing methods. |
Keywords
» Artificial intelligence » Generalization » Machine learning
