Summary of Machine Unlearning Through Fine-grained Model Parameters Perturbation, by Zhiwei Zuo et al.
Machine unlearning through fine-grained model parameters perturbation
by Zhiwei Zuo, Zhuo Tang, Kenli Li, Anwitaman Datta
First submitted to arxiv on: 9 Jan 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 This paper proposes novel machine unlearning techniques to protect user privacy by retracting data records from trained models, which can be computationally expensive. The authors suggest fine-grained approaches, Top-K and Random-k, that perturb model parameters to balance privacy needs with computational costs. These strategies aim to reduce the influence of sensitive data on trained models. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary Machine learning researchers have found ways to protect user privacy by removing or reducing data from trained models. This process is called machine unlearning. However, making sure this process doesn’t hurt the model’s performance can be tricky. The authors of this paper suggest new and efficient ways to do machine unlearning that balance privacy with computation. |
Keywords
* Artificial intelligence * Machine learning
