• Corpus ID: 257038524

Efficient Wireless Federated Learning with Partial Model Aggregation

  title={Efficient Wireless Federated Learning with Partial Model Aggregation},
  author={Zhixiong Chen and Wenqiang Yi and Arumugam Nallanathan and Geoffrey Y. Li},
The data heterogeneity across devices and the limited communication resources, e.g., bandwidth and energy, are two of the main bottlenecks for wireless federated learning (FL). To tackle these challenges, we first devise a novel FL framework with partial model aggregation (PMA). This approach aggregates the lower layers of neural networks, responsible for feature extraction, at the parameter server while keeping the upper layers, responsible for complex pattern recognition, at devices for… 

Figures and Tables from this paper



Efficient Federated Learning Algorithm for Resource Allocation in Wireless IoT Networks

A convergence upper bound is provided characterizing the tradeoff between convergence rate and global rounds, showing that a small number of active UEs per round still guarantees convergence and advocating the proposed FL algorithm for a paradigm shift in bandwidth-constrained learning wireless IoT networks.

Cost-Effective Federated Learning in Mobile Edge Networks

This paper analyzes how to design adaptive FL in mobile edge networks that optimally chooses these essential control variables to minimize the total cost while ensuring convergence, and develops a low-cost sampling-based algorithm to learn the convergence related unknown parameters.

Efficient Federated Meta-Learning Over Multi-Access Wireless Networks

This paper rigorously analyzes the contribution of each device to the global loss reduction in each round and develops an FML algorithm with a non-uniform device selection scheme to accelerate the convergence and formulate a resource allocation problem integrating NUFM in multi-access wireless systems to jointly improve the convergence rate.

Energy Efficient Federated Learning Over Wireless Communication Networks

An iterative algorithm is proposed where, at every step, closed-form solutions for time allocation, bandwidth allocation, power control, computation frequency, and learning accuracy are derived and can reduce up to 59.5% energy consumption compared to the conventional FL method.

Joint Device Scheduling and Resource Allocation for Latency Constrained Wireless Federated Learning

Experiments show that the proposed joint device scheduling and resource allocation policy to maximize the model accuracy within a given total training time budget for latency constrained wireless FL outperforms state-of-the-art scheduling policies under extensive settings of data distributions and cell radius.

Dynamic Scheduling for Heterogeneous Federated Learning in Private 5G Edge Networks

A dynamic scheduling algorithm (DISCO) is proposed, to make an intelligent decision on the set and order of scheduled devices in each communication round, and theoretical analysis reveals that under certain conditions, the learning performance and energy constraints can be guaranteed in the DISCO.

A Joint Learning and Communications Framework for Federated Learning Over Wireless Networks

Simulation results show that the proposed joint federated learning and communication framework can improve the identification accuracy by up to 1.4%, 3.5% and 4.1%, respectively, compared to an optimal user selection algorithm with random resource allocation and a wireless optimization algorithm that minimizes the sum packet error rates of all users while being agnostic to the FL parameters.

Dynamic Scheduling for Over-the-Air Federated Edge Learning With Energy Constraints

This work considers an over-the-air FEEL system with analog gradient aggregation, and proposes an energy-aware dynamic device scheduling algorithm to optimize the training performance within the energy constraints of devices, where both communication energy for gradient aggregation and computation energy for local training are considered.

Client Selection and Bandwidth Allocation in Wireless Federated Learning Networks: A Long-Term Perspective

A stochastic optimization problem for joint client selection and bandwidth allocation under long-term client energy constraints is formulated, and a new algorithm is developed that utilizes only currently available wireless channel information but can achieve long- term performance guarantee.

Decentralized Federated Learning With Unreliable Communications

  • Hao YeL. LiangG. Li
  • Computer Science
    IEEE Journal of Selected Topics in Signal Processing
  • 2022
It is proved that the proposed decentralized training system, even with unreliable communications, can still achieve the same asymptotic convergence rate as vanilla decentralized SGD with perfect communications.