Dynamic Participation in Federated Learning: Benchmarks and a Knowledge Pool Plugin
Not on the map. Why?
Abstract (arXiv)
Federated learning (FL) enables clients to collaboratively train a shared model in a distributed manner, setting it apart from traditional deep learning paradigms. However, most existing FL research assumes consistent client participation, overlooking the practical scenario of dynamic participation (DPFL), where clients may intermittently join or leave during training. Moreover, no existing benchmarking framework systematically supports the study of DPFL-specific challenges. In this work, we present the first open-source framework explicitly designed for benchmarking FL models under dynamic client participation. Our framework provides configurable data distributions, participation patterns, and evaluation metrics tailored to DPFL scenarios. Using this platform, we benchmark four major categories of widely adopted FL models and uncover substantial performance degradation under dynamic participation. To address these challenges, we further propose Knowledge-Pool Federated Learning (KPFL), a generic plugin that maintains a shared knowledge pool across both active and idle clients. KPFL leverages dual-age and data-bias weighting, combined with generative knowledge distillation, to mitigate instability and prevent knowledge loss. Extensive experiments demonstrate the significant impact of dynamic participation on FL performance and the effectiveness of KPFL in improving model robustness and generalization.
Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 15 of its 29 references that are in the map. First posted on arXiv 20 Nov 2025.
Builds on
15 of 29 references.
References · 15
- 2015Deep Residual Learning for Image Recognitioncited by 52,096
- 2016Communication-Efficient Learning of Deep Networks from Decentralized Datacited by 7,669
- 2018Federated Optimization in Heterogeneous Networkscited by 2,987
- 2021Model-Contrastive Federated Learningcited by 766
- 2013An Empirical Investigation of Catastrophic Forgetting in Gradient-Based Neural Networkscited by 722
- 2021Federated Learning on Non-IID Data Silos: An Experimental Studycited by 568
- 2019FedMD: Heterogenous Federated Learning via Model Distillationcited by 459
- 2021Data-Free Knowledge Distillation for Heterogeneous Federated Learningcited by 383
- 2020FedML: A Research Library and Benchmark for Federated Machine Learningcited by 300
- 2022Fine-tuning Global Model via Data-Free Knowledge Distillation for Non-IID Federated Learningcited by 172
- 2021Asynchronous Federated Learning on Heterogeneous Devices: A Surveycited by 93
- 2021Fast Federated Learning in the Presence of Arbitrary Device Unavailabilitycited by 72
3 more
- 2020Towards Flexible Device Participation in Federated Learningcited by 65
- 2021A distillation-based approach integrating continual learning and federated learning for pervasive servicescited by 42
- 2023TimelyFL: Heterogeneity-aware Asynchronous Federated Learning with Adaptive Partial Trainingcited by 27
Built on by
Not cited yet.
Similar, not cited
Close in wording, no citation either way
- 2024Harnessing Increased Client Participation with Cohort-Parallel Federated Learning
- 2024Knowledge Distillation in Federated Learning: a Survey on Long Lasting Challenges and New Solutions
- 2022Closing the Gap between Client and Global Model Performance in Heterogeneous Federated Learning
- 2023FedAL: Black-Box Federated Knowledge Distillation Enabled by Adversarial Learning
- 2025SFedKD: Sequential Federated Learning with Discrepancy-Aware Multi-Teacher Knowledge Distillation
- 2022Knowledge Distillation for Federated Learning: a Practical Guide
- 2022Handling Data Heterogeneity in Federated Learning via Knowledge Distillation and Fusion
- 2025Reviving Stale Updates: Data-Free Knowledge Distillation for Asynchronous Federated Learning