Heterogeneous clients with varying computing capabilities

The gap

Evidence profile

Stated in the inline gaps and conclusions and cells future research and cells research gap sections of the source papers, classified as general, spanning 4 journals.

Supporting evidence — 5 representative gaps

• Deep heterogeneity learning for cross-city transit forecasting: a differentially private federated framework with mixture-of-experts and seasonal decomposition (2026) · Frontiers in Future Transportation · doi

  Future research will explore adaptive DP mechanisms that dynamically adjust noise levels based on data sensitivity, model convergence, or specific client contributions, aiming for greater efficiency and tighter privacy guarantees with minimal accuracy sacrifice. Future work could focus on developing XAI techniques specific to federated MoE models to understand how different experts contribute to predictions for specific cities or scenarios, and how DP impacts interpretability. • Exploring Advanced Federated Personalization Strategies: While MoE provides adaptation, more explicit personalized federated learning (pFL) techniques, such as FedMeta, FedAvgM, or client clustering, could be investigated.

  generalstated in inline gapsevidence 5/5
  Keywords: specific federated future client techniques explore adaptive mechanisms dynamically adjust noise levels based sensitivity model
• Deep heterogeneity learning for cross-city transit forecasting: a differentially private federated framework with mixture-of-experts and seasonal decomposition (2026) · Frontiers in Future Transportation · doi

  Future research will explore adaptive DP mechanisms that dynamically adjust noise levels based on data sensitivity, model convergence, or specific client contributions, aiming for greater efficiency and tighter privacy guarantees with minimal accuracy sacrifice. Future work could focus on developing XAI techniques specific to federated MoE models to understand how different experts contribute to predictions for specific cities or scenarios, and how DP impacts interpretability. • Exploring Advanced Federated Personalization Strategies: While MoE provides adaptation, more explicit personalized federated learning (pFL) techniques, such as FedMeta, FedAvgM, or client clustering, could be investigated.

  generalstated in conclusionsevidence 5/5
  Keywords: specific federated future client techniques explore adaptive mechanisms dynamically adjust noise levels based sensitivity model
• A Hybrid Deep Learning Framework for Real-Time Network Intrusion Detection in IoT Environments (2026) · WSEAS TRANSACTIONS ON COMPUTER RESEARCH · doi

  To explore semisupervised and federated learning methods to reduce labeled data and ensure privacy in collaborative IoT contexts. - To evaluate the performance of the proposed model in real-world scenarios. - To improve the robustness and generalizability of the proposed model.

  generalstated in cells future researchevidence 5/5
  Keywords: explore semisupervised federated learning methods reduce labeled data
• FedSplitX: Federated split learning for computationally-constrained heterogeneous clients (2026) · Neurocomputing · doi

  Heterogeneous clients with varying computing capabilities pose challenges in federated learning. - Prior techniques have limitations in handling heterogeneous clients. - There is a need for a novel federated learning framework that can tackle system heterogeneity.

  generalstated in cells research gapevidence 5/5
  Keywords: heterogeneous clients varying computing capabilities pose challenges federated
• Resilient federated learning under data and system heterogeneity via genetic algorithm-based group client selection (2026) · Evolutionary Intelligence · doi

  Client selection is a critical challenge in federated learning due to non-IID heterogeneity in data distributions and heterogeneity in computational resources. - Prior work has addressed client selection using individual-based approaches, but these approaches ignore group effects.

  generalstated in cells research gapevidence 5/5
  Keywords: client selection critical challenge federated learning due non-iid

Questions about this gap