Summary of What Can We Learn From State Space Models For Machine Learning on Graphs?, by Yinan Huang et al.
What Can We Learn from State Space Models for Machine Learning on Graphs?
by Yinan Huang, Siqi Miao, Pan Li
First submitted to arxiv on: 9 Jun 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 paper proposes Graph State Space Convolution (GSSC), a novel approach to extend State Space Models (SSMs) to graph-structured data. GSSC leverages global permutation-equivariant set aggregation and factorizable graph kernels that rely on relative node distances as the convolution kernels, preserving the strengths of SSMs. The authors demonstrate the provably stronger expressiveness of GSSC than Message Passing Neural Networks (MPNNs) in counting graph substructures and show its effectiveness across 11 real-world benchmark datasets. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary GSSC is a new way to do machine learning on graphs, which are important for things like social networks and molecular structures. Right now, we have two main ways to do this: Message Passing Neural Networks (MPNNs) and Graph Transformers. MPNNs are good at some things, but they’re not very powerful. Graph Transformers can be really powerful, but they use up a lot of computer resources. The authors of this paper found a way to make something called State Space Models (SSMs) work on graphs, which is good because SSMs are efficient and can handle long sequences of data. They used the strengths of two other models, Recurrent Neural Networks (RNNs) and Convolutional Neural Networks (CNNs), to create this new approach. |
Keywords
» Artificial intelligence » Machine learning
