Summary of Modeling Inter-intra Heterogeneity For Graph Federated Learning, by Wentao Yu et al.
Modeling Inter-Intra Heterogeneity for Graph Federated Learning
by Wentao Yu, Shuo Chen, Yongxin Tong, Tianlong Gu, Chen Gong
First submitted to arxiv on: 16 Dec 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 novel Federated learning (Fed) approach, named Federated learning by integrally modeling the Inter-Intra Heterogeneity (FedIIH), to address the issue of heterogeneity in graph data. The authors argue that existing methods rely on unreliable inter-subgraph similarities estimated from local models and ignore intra-heterogeneity within each subgraph. FedIIH addresses these issues by inferring whole distribution of subgraph data using hierarchical variational models for inter-subgraph relationships, and disentangling subgraphs into latent factors to learn robust representations. The method considers both inter- and intra-heterogeneity simultaneously and is evaluated on six homophilic and five heterophilic graph datasets in both non-overlapping and overlapping settings. FedIIH outperforms nine state-of-the-art methods by a large margin of 5.79% on all heterophilic datasets. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This paper develops a new way to connect different graphs together using artificial intelligence (AI). Graphs are like maps that show connections between things, and connecting them is important for many applications. The problem is that these graphs can be very different from each other, which makes it hard to combine them. The authors propose a new method called FedIIH that takes into account both the differences between the graphs and the differences within each graph. They test their method on many real-world datasets and show that it performs much better than existing methods. |
Keywords
» Artificial intelligence » Federated learning
