Dynamic Participation in Federated Learning: Benchmarks and a Knowledge Pool Plugin

TL;DR

Dynamic Participation in Federated Learning: Benchmarks and a Knowledge Pool Plugin presents the first open-source framework to benchmark FL under dynamic client participation (DPFL). It defines DPFL data heterogeneity using Dirichlet distributions, participation dynamics via Timed-Random and Markovian patterns, and DPFL-specific metrics to quantify robustness and stability. The work benchmarks nine FL models across four families and introduces KPFL, a dual-age, data-bias weighted knowledge pool with generative knowledge distillation, to mitigate DPFL-induced degradation. Extensive experiments on the Office-Caltech dataset demonstrate that DPFL significantly harms existing methods, while KPFL improves robustness and generalization and scales to larger pools, validating its broad applicability across FL models.

Abstract

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.

Figures (2)

• Figure 1: The proposed DPFL benchmarking framework.
• Figure 2: The Knowledge-Pool Federated Learning framework.