Abstract:In this work, a <italic>Hybrid Hierarchical Federated Edge Learning</italic> (HHFEL) architecture that consists of a device layer, an edge layer, and a cloud layer over heterogeneous networks, is investigated for large-scale model training. In such systems, learning efficiency is severely degraded by limited communication resources and device heterogeneity in terms of local data distribution and computation capability, especially for synchronous FL mechanisms where the training of each round should wait for the slowest device. To tackle this issue, asynchronous FL is proposed, which allows the devices with powerful computation and communication capabilities exchanging information with the server more frequently. However, this asynchronous FL framework faces a new challenge of low accuracy caused by the imbalanced local model updating. To overcome the shortage of both synchronous and asynchronous FLs, we propose an enhanced online semi-asynchronous FL mechanism between the edge-device layers, where each device trains its local model with the newly generated data and each edge server aggregates a number of local models based on their arrival order in each round. Particularly, devices with faster training speeds would fully utilize the idle time by training their local models repetitively. Meanwhile, synchronous FL with an edge elastic update strategy is adopted to the cloud-edge layers for personalized information exchange. Considering the continuous data generation feature, we formulate the objective problem as an online <italic>Markov Decision Process</italic> (MDP) to realize efficient communication-and-computing HHFEL via joint device selection and resource allocation. Due to the non-convex and combinatorial problem structure, we develop a hybrid <italic>Deep Q-Network</italic> (DQN) and <italic>Deep Deterministic Policy Gradient</italic> (DDPG) approach with low computational complexity to adapt the device selection and resource allocation strategies. Numerical results show the effectiveness of the proposed mechanism compared with existing benchmarks.

Study of Hierarchical Federation Architecture of Submarine Voyage Training Simulation System

Automatic construction of Hierarchical Federations Architecture

An extended HLA multilayer federation integration architecture for multidisciplinary collaborative simulation

HLA-based Prototype Design and Development of Environment Modeling and Simulation for Deep Sea Mining

A Hierarchical Federated Integration Architecture For Collaborative Product Development

Research on Multi-disciplinary Modeling Based on HLA

A collaborative platform for complex product design with an extended HLA integration architecture

Collaborative Development System Based on Federation Integration

Research of HLA-Based Multi-Federation Collaborative Simulation Management System for Complex Product Design

Research on Extended HLA Technique Based on Meta-MDA

Design and Implementation of Submarine Warfare Simulation System Based on EJB and HLA

Bridging Technology for Composing Large-scale HLA Federation

Efficient communication approach in HLA-based distributed simulation for VR application

Ontology based collaborative simulation framework using HLA & Web services

To Implement Distributed Multidisciplinary Federated Simulation for Collaborative Product Development

On HLA-Based Collaborative Simulation Techniques.

Research of HLA/DEVS-based High Level Modeling for Collaborative Simulation

Multidisciplinary Simulation Of Space-Based Telescope Using Hla

Research on Simulation System for Ship Transport Operation Based on HLA

Enhanced Hybrid Hierarchical Federated Edge Learning Over Heterogeneous Networks