mlsanityData to 14 Oct 2026

Joint Routing and Model Pruning for Decentralized Federated Learning in Bandwidth-Constrained Multi-Hop Wireless Networks

arXiv 2603.15188 ↗First posted 16 Mar 2026cs.LG cs.NIcited by 0

Not on the map. Why?

Abstract (arXiv)

Decentralized federated learning (D-FL) enables privacy-preserving training without a central server, but multi-hop model exchanges and aggregation are often bottlenecked by communication resource constraints. To address this issue, we propose a joint routing-and-pruning framework that optimizes routing paths and pruning rates to maintain communication latency within prescribed limits. We analyze how the sum of model biases across all clients affects the convergence bound of D-FL and formulate an optimization problem that maximizes the model retention rate to minimize these biases under communication constraints. Further analysis reveals that each client's model retention rate is path-dependent, which reduces the original problem to a routing optimization. Leveraging this insight, we develop a routing algorithm that selects latency-efficient transmission paths, allowing more parameters to be delivered within the time budget and thereby improving D-FL convergence. Simulations demonstrate that, compared with unpruned systems, the proposed framework reduces average transmission latency by 27.8% and improves testing accuracy by approximately 12%. Furthermore, relative to standard benchmark routing algorithms, the proposed routing method improves accuracy by roughly 8%.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 21 of its 63 references that are in the map. First posted on arXiv 16 Mar 2026.

Builds on

21 of 63 references.

References · 21

  1. 2016Communication-Efficient Learning of Deep Networks from Decentralized Datacited by 7,669
  2. 2020HeteroFL: Computation and Communication Efficient Federated Learning for Heterogeneous Clientscited by 337
  3. 2019Model Pruning Enables Efficient Federated Learning on Edge Devicescited by 222
  4. 2023A Survey on Deep Neural Network Pruning-Taxonomy, Comparison, Analysis, and Recommendationscited by 202
  5. 2021Adaptive Quantization of Model Updates for Communication-Efficient Federated Learningcited by 63
  6. 2021Decentralized Federated Learning: Balancing Communication and Computing Costscited by 55
  7. 2021Federated Learning over Wireless Device-to-Device Networks: Algorithms and Convergence Analysiscited by 51
  8. 2020Multi-Stage Hybrid Federated Learning over Large-Scale D2D-Enabled Fog Networkscited by 42
  9. 2020Decentralized Federated Learning via SGD over Wireless D2D Networkscited by 33
  10. 2023Federated Domain Generalization: A Surveycited by 25
  11. 2021FedDQ: Communication-Efficient Federated Learning with Descending Quantizationcited by 20
  12. 2023Can Unstructured Pruning Reduce the Depth in Deep Neural Networks?cited by 20
9 more
  1. 2021Federated Learning Over Wireless Channels: Dynamic Resource Allocation and Task Schedulingcited by 8
  2. 2022Efficient Wireless Federated Learning with Partial Model Aggregationcited by 8
  3. 2023Delay Sensitive Hierarchical Federated Learning with Stochastic Local Updatescited by 8
  4. 2023Multi-Source to Multi-Target Decentralized Federated Domain Adaptationcited by 6
  5. 2019On-Demand Routing for Urban VANETs using Cooperating UAVscited by 3
  6. 2023Tram-FL: Routing-based Model Training for Decentralized Federated Learningcited by 3
  7. 2024Quantized and Asynchronous Federated Learningcited by 3
  8. 2023CommunityAI: Towards Community-based Federated Learningcited by 2
  9. 2025Route-and-Aggregate Decentralized Federated Learning Under Communication Errorscited by 2

Built on by

Not cited yet.