Summary of Homomorphism Counts As Structural Encodings For Graph Learning, by Linus Bao et al.
Homomorphism Counts as Structural Encodings for Graph Learning
by Linus Bao, Emily Jin, Michael Bronstein, İsmail İlkan Ceylan, Matthias Lanzinger
First submitted to arxiv on: 24 Oct 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 Graph Transformers have been widely adopted to process graph-structured data. This paper proposes a new structural encoding framework called Motif Structural Encoding (MoSE) that counts graph homomorphisms to condition the model on graph structure. MoSE is compared theoretically to other well-known encodings, including random-walk structural encoding, and shown to have similar expressive power to standard message passing neural networks. Empirically, MoSE outperforms these encodings across various architectures, achieving state-of-the-art performance on a molecular property prediction dataset. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This paper is about improving how computer programs process data that has connections or relationships between things. It introduces a new way to do this called Motif Structural Encoding (MoSE). MoSE helps the program understand these connections by counting patterns in the data. This new approach works better than other ways of doing this and can be used for important tasks like predicting properties of molecules. |
