Summary of Multi-source Unsupervised Domain Adaptation on Graphs with Transferability Modeling, by Tianxiang Zhao and Dongsheng Luo and Xiang Zhang and Suhang Wang
Multi-source Unsupervised Domain Adaptation on Graphs with Transferability Modeling
by Tianxiang Zhao, Dongsheng Luo, Xiang Zhang, Suhang Wang
First submitted to arxiv on: 14 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 This paper tackles the problem of multi-source unsupervised domain adaptation (MSUDA) for graphs, aiming to train models on annotated source domains that can be transferred to unsupervised target graphs for node classification. The key challenge is selecting good source instances and adapting the model, which previous approaches found less effective due to diverse graph structures. The proposed framework, Selective Multi-source Adaptation for Graph (SMAG), uses a domain selector, sub-graph node selector, and bi-level alignment objective for adaptation. The framework first disentangles similarity across graphs by measuring transferability of a graph-modeling task set, using it as evidence for source domain selection. A node selector is then incorporated to capture variation in transferability within the same source domain. To learn invariant features for adaptation, SMAG aligns the target domain to selected source data at both embedding space and classification level. The proposed method achieves significant results on five graph datasets, demonstrating its effectiveness. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary In simple terms, this paper explores how machines can adapt from one type of graph information to another without being specifically taught. It’s like teaching a machine to recognize a new type of image just by looking at similar images it has learned before. The researchers developed a special method called SMAG that helps the machine learn what matters and what doesn’t, allowing it to make better predictions on new data. |
Keywords
» Artificial intelligence » Alignment » Classification » Domain adaptation » Embedding space » Transferability » Unsupervised
