Summary of Dynamic Graph Unlearning: a General and Efficient Post-processing Method Via Gradient Transformation, by He Zhang et al.
Dynamic Graph Unlearning: A General and Efficient Post-Processing Method via Gradient Transformation
by He Zhang, Bang Wu, Xiangwen Yang, Xingliang Yuan, Xiaoning Liu, Xun Yi
First submitted to arxiv on: 23 May 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 Dynamic graph neural networks (DGNNs) have been widely deployed in web applications for personalized content delivery due to their ability to learn from complex user interaction data. However, users have raised privacy concerns regarding the misuse of personal data for model training, requiring DGNNs to “forget” their data to meet AI governance laws such as the “right to be forgotten” in GDPR. Despite existing static graph unlearning studies, there is a need for dynamic graph unlearning methods that can effectively and efficiently implement this process. This paper proposes an effective, efficient, general, and post-processing method called Gradient Transformation that directly maps the unlearning request to the desired parameter update. Evaluations on six real-world datasets and state-of-the-art DGNN backbones demonstrate its effectiveness in reducing utility loss while maintaining speed and performance gains. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary Imagine you use a website that knows what you like and shows you personalized content. But, you might worry about your personal data being used for something else without your permission. To solve this problem, researchers have developed a way to “forget” the data collected by these websites. This paper presents a new method called Gradient Transformation that can do this efficiently while still providing good results. The method was tested on six different datasets and showed promising results. |
