# Optimizing Split Federated Learning with Unstable Client Participation

arXiv: https://arxiv.org/abs/2509.17398 · HTML: https://mlsanity.com/paper/2509.17398/

## Key facts

- Cited by 5 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 22 Sep 2025.

## Built on by (most cited first, 5 of 5)

- [LLM-Driven Stationarity-Aware Expert Demonstrations for Multi-Agent Reinforcement Learning in Mobile Systems](https://mlsanity.com/paper/2511.19368/) (2025)
- [SIDeR: Semantic Identity Decoupling for Unrestricted Face Privacy](https://mlsanity.com/paper/2602.04994/) (2026)
- [Efficient Resource Optimization for Split Federated Learning](https://mlsanity.com/paper/2608.17849/) (2026)
- [SeaCausal-FL: Federated Fuzzy Causal Learning for Maritime IoT Fault Diagnosis and Counterfactual Reasoning](https://mlsanity.com/paper/2609.06257/) (2026)
- [Redundancy Meets Synergy: Dependency-aware Expert Selection for MoE via Submodular Optimization](https://mlsanity.com/paper/2610.00558/) (2026)

## Abstract (arXiv)

To enable training of large artificial intelligence (AI) models at the network edge, split federated learning (SFL) has emerged as a promising approach by distributing computation between edge devices and a server. However, while unstable network environments pose significant challenges to SFL, prior schemes often overlook such an effect by assuming perfect client participation, rendering them impractical for real-world scenarios. In this work, we develop an optimization framework for SFL with unstable client participation. We theoretically derive the first convergence upper bound for SFL with unstable client participation by considering activation uploading failures, gradient downloading failures, and model aggregation failures. Based on the theoretical results, we formulate a joint optimization problem for client sampling and model splitting to minimize the upper bound. We then develop an efficient solution approach to solve the problem optimally. Extensive simulations on EMNIST and CIFAR-10 demonstrate the superiority of our proposed framework compared to existing benchmarks.