Summary of Mixture Of Link Predictors on Graphs, by Li Ma et al.
Mixture of Link Predictors on Graphs
by Li Ma, Haoyu Han, Juanhui Li, Harry Shomer, Hui Liu, Xiaofeng Gao, Jiliang Tang
First submitted to arxiv on: 13 Feb 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 The proposed Link-MoE model is a significant advancement in link prediction, which aims to forecast unseen connections in graphs. By recognizing that different node pairs within the same dataset require varied pairwise information for accurate prediction, Link-MoE utilizes various Graph Neural Networks (GNNs) as experts and strategically selects the appropriate expert for each node pair based on various types of pairwise information. This approach leads to substantial performance improvements compared to state-of-the-art baselines, with relative gains of 18.71% on the MRR metric for the Pubmed dataset and 9.59% on the Hits@100 metric for the ogbl-ppa dataset. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary Link prediction is a way to find new connections between things in a graph. This is important because graphs are used to represent relationships between different types of data. The problem with current methods is that they don’t work well when different parts of the graph need different information to make predictions. To fix this, researchers developed a new model called Link-MoE. This model uses many smaller models (called experts) and chooses which one to use based on the type of information needed for each prediction. By doing this, Link-MoE is able to make much more accurate predictions than previous methods. |
