2025· Neural Information Processing Systems· 2 citations· 55 references
Computer Science
TL;DR
This work proposes a FedIGL framework based on invariant learning, which effectively disrupts spurious correlations and further mines the invariant factors across different distributions, and proposes a novel Bi-Gradient Regularization strategy that introduces gradient constraints to guide the model in identifying client-agnostic and client-specific subgraph patterns for better graph representations.
Abstract
Federated Graph Learning (FGL) effectively facilitates cross-domain graph model training by enabling decentralized learning across multiple domains, while ensuring data privacy through local data storage and communication of model updates instead of raw data. Existing approaches usually assume shared generic knowledge (e.g., prototypes, spectral features) via aggregating local structures statistically to alleviate structural heterogeneity. However, imposing overly strict assumptions about the presumed correlation between structural features and the global objective often fails in generalizing to local tasks, leading to suboptimal performance. To tackle this issue, we propose a Fed erated I nvariant G raph L earning ( FedIGL ) framework based on invariant learning, which effectively disrupts spurious correlations and further mines the invariant factors across different distributions. Specifically, a server-side global model is trained to capture client-agnostic subgraph patterns shared across clients, whereas client-side models specialize in client-specific sub-graph patterns. Subsequently, without compromising privacy, we propose a novel Bi-Gradient Regularization strategy that introduces gradient constraints to guide the model in identifying client-agnostic and client-specific subgraph patterns for better graph representations. Extensive experiments on graph-level clustering and classification tasks demonstrate the superiority of FedIGL against its competitors.
FedTopo is proposed, a relation-level framework that encodes global knowledge as class relation topology, capturing how classes relate within each client rather than where they lie in feature space.
Zhaoyang Ma, Zhihao Wu, Xin Gao et al.· 0 citations
Federated learning enables privacy-preserving collaborative training, but highly heterogeneous client data remain challenging, especially in graph federated learning where clients possess structurally diverse graphs. Existing personalized federated learning (PFL) methods ignore the intrinsic geometric properties of diverse graph structures. We propose FlatLand, a novel personalized federated learning method that embeds different clients'data in tailored Lorentz space of hyperbolic geometry. Our key insight is that hyperbolic geometry naturally accommodates the intrinsic negative curvature prevalent in real-world graphs, while the time-like dimension in Lorentz space provides a principled way to encode client-specific heterogeneity. We develop a parameter decoupling strategy that separates heterogeneous information (captured in time-like parameters) from common knowledge (preserved in space-like parameters), enabling direct aggregation without requiring client similarity estimation and extra calculation modules. Empirical results on diverse federated graph learning tasks demonstrate that FlatLand achieves superior performance, particularly in low-dimensional settings.
Jiahong Liu, Ram Samarth, Xinyu Fu et al.· 0 citations
Subgraph federated learning (subgraph FL) enables collaborative graph neural network training without sharing raw graph data, but suffers from severe Non-IID distributions and structural fragmentation. In such settings, Non-IID distributions induce pronounced client specialization, where each client excels in a subset of nodes but remains insufficiently trained on others. However, existing similarity-based aggregation and distillation methods fail to balance client specialization and global generalization, often reinforcing dominant local representations and impairing generalization on under-represented nodes, or over-incorporating global knowledge and thereby disrupting well-optimized local specialization. In this paper, we propose FedDUA, a novel disagreement-aware and uncertainty-guided framework for subgraph FL. Specifically, FedDUA first models cross-client semantic disagreement via a lightweight semantic anchor graph and derives adaptive aggregation weights for reliable global federated knowledge. On the client side, FedDUA introduces an uncertainty-aware local-global semantic alignment mechanism that selectively reinforces representations of confident nodes while guiding uncertain nodes with aggregated global knowledge, thereby balancing local specialization and global generalization. Extensive experiments on six real-world datasets demonstrate that FedDUA consistently outperforms the state-of-the-art subgraph FL methods across varying numbers of clients. Further analyses validate the robustness and effectiveness of the proposed disagreement modeling and uncertainty-aware local-global semantic alignment strategies.
Keao Xi, Nannan Wu, Yiming Zhao et al.· Proceedings of the 32nd ACM...· 0 citations
DC-FNCD (Divide-and-Conquer based Federated Nonlinear Causal Discovery) is proposed, which decomposes the global problem into independent per-variable neighborhood learning tasks, enabling lossless federated aggregation without raw data exchange.
Xianjie Guo, Shuai Yang, Lin Ma et al.· Proceedings of the 32nd ACM...· 0 citations
A domain-sensitive federated pruning framework that preserves domain-invariant structures while retaining domain-specific representations and a structure-aware aggregation algorithm that fuses heterogeneous personalized architectures into a domain-generalized global model is proposed.
Chenchen Lin, Wenhao Yuan, Zhengji Xu et al.· 0 citations