aiqbench analysis
FedTopo addresses the critical challenge of knowledge transfer in federated learning environments where client models possess heterogeneous architectures. By shifting from absolute feature space alignment to relation-level topology sharing, this method effectively bypasses representation misalignment issues that typically degrade performance across diverse backbones. The approach demonstrates robust efficacy with minimal communication overhead and no inference cost, offering a practical solution for scalable decentralized training systems.
FedTopo: Relation-Level Topology Sharing for Model-Heterogeneous Federated Learning
arXiv:2607.26801v1 Announce Type: cross Abstract: Federated learning (FL) enables collaborative learning over decentralized data silos without centralizing raw data. However, heterogeneous local architectures often induce non-aligned representation spaces, making it difficult to transfer global knowledge across silos. Existing paradigms share this knowledge as model parameters, distilled predictions, or class prototypes, yet all encode it in an absolute space that must be aligned across clients. Heterogeneous backbones break this alignment, so the shared knowledge becomes unreliable and misleads local training. We propose FedTopo, 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. Each client builds its relation topology from local prototypes and uploads it with class statistics. The server then aggregates these relations in a reliability-aware manner that down-weights weakly supported ones, and broadcasts the global topology to clients. The global topology guides local training by emphasizing topology-similar negative classes. Experiments on three datasets under eight heterogeneous backbones show that FedTopo consistently outperforms parameter-, distillation-, and prototype-sharing baselines, with low communication and no inference overhead. Our code is available at https://github.com/Zhaoyang-Ma/FedTopo.
Abstract:Federated learning (FL) enables collaborative learning over decentralized data silos without centralizing raw data. However, heterogeneous local architectures often induce non-aligned representation spaces, making it difficult to transfer global knowledge across silos. Existing paradigms share this knowledge as model parameters, distilled predictions, or class prototypes, yet all encode it in an absolute space that must be aligned across clients. Heterogeneous backbones break this alignment, so the shared knowledge becomes unreliable and misleads local training. We propose FedTopo, 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. Each client builds its relation topology from local prototypes and uploads it with class statistics. The server then aggregates these relations in a reliability-aware manner that down-weights weakly supported ones, and broadcasts the global topology to clients. The global topology guides local training by emphasizing topology-similar negative classes. Experiments on three datasets under eight heterogeneous backbones show that FedTopo consistently outperforms parameter-, distillation-, and prototype-sharing baselines, with low communication and no inference overhead. Our code is available at this https URL.
13 pages, 9 figures, 4 tables. Submitted to ICDE 2027 (first submission round)
Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
arXiv:2607.26801 [cs.LG]
(or arXiv:2607.26801v1 [cs.LG] for this version)
https://doi.org/10.48550/arXiv.2607.26801 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Full text under the publisher's open licence · 294 words
Related coverage
News
Quantum Computing Report · Jul 25, 2026
Spur: Modern GPU Job Scheduling for HPC and AI Workloads
AMD · Jul 22, 2026
FedIncome: Federated Learning for Income Estimation in Digital Lending Under Data Sovereignty Constraints
arXiv · Sep 24, 2026
AI by Default
LessWrong · Oct 17, 2026