Virtual Nodes Can Help: Tackling Distribution Shifts in Federated Graph Learning
Not on the map. Why?
Abstract (arXiv)
Federated Graph Learning (FGL) enables multiple clients to jointly train powerful graph learning models, e.g., Graph Neural Networks (GNNs), without sharing their local graph data for graph-related downstream tasks, such as graph property prediction. In the real world, however, the graph data can suffer from significant distribution shifts across clients as the clients may collect their graph data for different purposes. In particular, graph properties are usually associated with invariant label-relevant substructures (i.e., subgraphs) across clients, while label-irrelevant substructures can appear in a client-specific manner. The issue of distribution shifts of graph data hinders the efficiency of GNN training and leads to serious performance degradation in FGL. To tackle the aforementioned issue, we propose a novel FGL framework entitled FedVN that eliminates distribution shifts through client-specific graph augmentation strategies with multiple learnable Virtual Nodes (VNs). Specifically, FedVN lets the clients jointly learn a set of shared VNs while training a global GNN model. To eliminate distribution shifts, each client trains a personalized edge generator that determines how the VNs connect local graphs in a client-specific manner. Furthermore, we provide theoretical analyses indicating that FedVN can eliminate distribution shifts of graph data across clients. Comprehensive experiments on four datasets under five settings demonstrate the superiority of our proposed FedVN over nine baselines.
Key facts. Cited by 3 papers in the map (data through 14 Oct 2026). Builds on 41 of its 53 references that are in the map. First posted on arXiv 26 Dec 2024.
Builds on
41 of 53 references.
References · 41
- 2020A Simple Framework for Contrastive Learning of Visual Representationscited by 12,049
- 2016Semi-Supervised Classification with Graph Convolutional Networkscited by 11,361
- 2017Graph Attention Networkscited by 8,923
- 2016Communication-Efficient Learning of Deep Networks from Decentralized Datacited by 7,669
- 2017Inductive Representation Learning on Large Graphscited by 6,320
- 2018How Powerful are Graph Neural Networks?cited by 4,393
- 2017Neural Message Passing for Quantum Chemistrycited by 3,487
- 2018Federated Optimization in Heterogeneous Networkscited by 2,987
- 2018Dynamic Graph CNN for Learning on Point Cloudscited by 2,223
- 2020Open Graph Benchmark: Datasets for Machine Learning on Graphscited by 2,163
- 2021Barlow Twins: Self-Supervised Learning via Redundancy Reductioncited by 1,838
- 2018Hierarchical Graph Representation Learning with Differentiable Poolingcited by 787
29 more
- 2016Automatic chemical design using a data-driven continuous representation of moleculescited by 765
- 2020Ditto: Fair and Robust Federated Learning Through Personalizationcited by 672
- 2021FedBN: Federated Learning on Non-IID Features via Local Batch Normalizationcited by 571
- 2021Exploiting Shared Representations for Personalized Federated Learningcited by 541
- 2020Parameterized Explainer for Graph Neural Networkcited by 426
- 2020Explainability in Graph Neural Networks: A Taxonomic Surveycited by 384
- 2021OGB-LSC: A Large-Scale Challenge for Machine Learning on Graphscited by 329
- 2021On Explainability of Graph Neural Networks via Subgraph Explorationscited by 281
- 2020Iterative Deep Graph Learning for Graph Neural Networks: Better and Robust Node Embeddingscited by 214
- 2022Learning Causally Invariant Representations for Out-of-Distribution Generalization on Graphscited by 144
- 2022Federated Learning from Pre-Trained Models: A Contrastive Learning Approachcited by 142
- 2022GOOD: A Graph Out-of-Distribution Benchmarkcited by 141
- 2021Subgraph Federated Learning with Missing Neighbor Generationcited by 132
- 2021Federated Graph Classification over Non-IID Graphscited by 114
- 2021SLAPS: Self-Supervision Improves Structure Learning for Graph Neural Networkscited by 110
- 2022Federated Learning on Non-IID Graphs via Structural Knowledge Sharingcited by 88
- 2022Graph Rationalization with Environment-based Augmentationscited by 70
- 2021Adaptive Graph Convolution for Point Cloud Analysiscited by 60
- 2022Towards Understanding and Mitigating Dimensional Collapse in Heterogeneous Federated Learningcited by 46
- 2022Unleashing the Power of Graph Data Augmentation on Covariate Distribution Shiftcited by 38
- 2024Federated Graph Semantic and Structural Learningcited by 30
- 2023Joint Learning of Label and Environment Causal Independence for Graph Out-of-Distribution Generalizationcited by 29
- 2021FedNI: Federated Graph Learning with Network Inpainting for Population-Based Disease Predictioncited by 26
- 2023Neighbor Contrastive Learning on Learnable Graph Augmentationcited by 25
- 2024Federated Graph Learning with Structure Proxy Alignmentcited by 14
- 2024CLDG: Contrastive Learning on Dynamic Graphscited by 13
- 2024Safety in Graph Machine Learning: Threats and Safeguardscited by 11
- 2023Spatial-Temporal Networks for Antibiogram Pattern Predictioncited by 10
- 2024Federated Graph Learning with Graphless Clientscited by 3
Built on by
Citing papers per year: 2025: 3.
3 papers.