Summary of Promoting Fairness in Link Prediction with Graph Enhancement, by Yezi Liu et al.
Promoting Fairness in Link Prediction with Graph Enhancement
by Yezi Liu, Hanning Chen, Mohsen Imani
First submitted to arxiv on: 13 Sep 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 This paper proposes a method called FairLink for fair link prediction in network analysis. The task of predicting links between nodes can be prone to biased predictions when links are unfairly predicted between nodes from different sensitive groups. Existing methods try to mitigate this issue by incorporating debiasing techniques within graph embeddings. However, training on large real-world graphs is already challenging, and adding fairness constraints can further complicate the process. FairLink learns a fairness-enhanced graph that bypasses the need for debiasing during the link predictor’s training. This method maintains link prediction accuracy while enhancing fairness by minimizing the absolute difference in link probabilities between node pairs within the same sensitive group and those between node pairs from different sensitive groups. The proposed approach is evaluated on multiple large-scale graphs, demonstrating that FairLink not only promotes fairness but also often achieves link prediction accuracy comparable to baseline methods. Additionally, the enhanced graph exhibits strong generalizability across different GNN architectures. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This paper helps solve a problem with predicting links between nodes in networks. When we predict links, we might unfairly favor or disfavor certain groups of nodes. The researchers propose a new way to do link prediction that is fair and unbiased. This method learns a special kind of graph that can be used for link prediction without introducing bias. They test their approach on many large networks and show that it works well and maintains its accuracy even when used with different types of machine learning models. |
Keywords
» Artificial intelligence » Gnn » Machine learning
