Asynchronous Federated Learning with Bidirectional Quantized Communications and Buffered Aggregation

2023/08/01 by Tomás Ortega del Rincón, Ortega, Tomas, Hamid Jafarkhani +1 · 2 citations
Computer Science · #68W10 #68W15 #68W40 #90C06 #90C26 #90C35 #Cooperative Communication and Network Coding #E.4 #F.2.1 #FOS: Computer and information sciences #FOS: Electrical engineering #FOS: Mathematics #G.1.6 #Machine Learning (cs.LG) #Optimization and Control (math.OC) #Privacy-Preserving Technologies in Data #Signal Processing (eess.SP) #Stochastic Gradient Optimization Techniques #electronic engineering #information engineering

paper · pdf · doi:10.48550/arxiv.2308.00263

openalex publication_date 2023/08/01 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28

Abstract

Asynchronous Federated Learning with Buffered Aggregation (FedBuff) is a state-of-the-art algorithm known for its efficiency and high scalability. However, it has a high communication cost, which has not been examined with quantized communications. To tackle this problem, we present a new algorithm (QAFeL), with a quantization scheme that establishes a shared "hidden" state between the server and clients to avoid the error propagation caused by direct quantization. This approach allows for high precision while significantly reducing the data transmitted during client-server interactions. We provide theoretical convergence guarantees for QAFeL and corroborate our analysis with experiments on a standard benchmark.

Cited by

• Communication Compression for Distributed Learning with Aggregate and Server-Guided Feedback
• Stragglers Can Contribute More: Uncertainty-Aware Distillation for Asynchronous Federated Learning

Related

• FedFQ: Federated Learning with Fine-Grained Quantization
• DAdaQuant: Doubly-adaptive quantization for communication-efficient\n Federated Learning
• Communication Efficient Federated Learning with Adaptive Quantization
• FedAQ: Communication-Efficient Federated Edge Learning via Joint Uplink and Downlink Adaptive Quantization
• Communication-Efficient Federated Learning via Clipped Uniform Quantization