Abstract:This paper presents an approximate wireless communication scheme for federated learning (FL) model aggregation in the uplink transmission. We consider a realistic channel that reveals bit errors during FL model exchange in wireless networks. Our study demonstrates that random bit errors during model transmission can significantly affect FL performance. To overcome this challenge, we propose an approximate communication scheme based on the mathematical and statistical proof that machine learning (ML) model gradients are bounded under certain constraints. This bound enables us to introduce a novel encoding scheme for float-to-binary representation of gradient values and their QAM constellation mapping. Besides, since FL gradients are error-resilient, the proposed scheme simply delivers gradients with errors when the channel quality is satisfactory, eliminating extensive error-correcting codes and/or retransmission. The direct benefits include less overhead and lower latency. The proposed scheme is well-suited for resource-constrained devices in wireless networks. Through simulations, we show that the proposed scheme is effective in reducing the impact of bit errors on FL performance and saves at least half the time than transmission with error correction and retransmission to achieve the same learning performance. In addition, we investigated the effectiveness of bit protection mechanisms in high-order modulation when gray coding is employed and found that this approach considerably enhances learning performance.

Approximation of Likelihood Function Based on Gaussian Model

Gaussian mixture density modeling and decomposition with weighted likelihood

Likelihood approximations via Gaussian approximate inference

A Novel Near Maximum Likelihood Detection Scheme for Wireless Mimo Systems

Approximation to Independent Lognormal Sum with Α–μ Distribution and the Application

Gaussian Mixture Model for Millimeter-Wave Cellular Communication Networks

Approximation to independent lognormal sum with α-µ distribution and the application

Approximate Optimal Estimation Based on Kullback–Leibler Divergence for Lossy Networks Without Acknowledgement

Approximate Distribution of Log 2 ( A + Χ 2 ) and Its Applications

Learning-Based Maximum Likelihood Estimator for Angle-of-Arrival Localization

An Approximation to the Likelihood Function for Band-Power Estimates of CMB Anisotropies

A generalized soft likelihood function in combining multi-source belief distribution functions

Hierarchical low rank approximation of likelihoods for large spatial datasets

A Localization Error Estimation Method Based on Maximum Likelihood for Wireless Sensor Networks

Maximum Approximated Likelihood Estimation

Generalized Maximum Likelihood Algorithm for Direction-of-Arrival Estimation of Coherent Sources

Beyond Gaussian Noise: A Generalized Approach to Likelihood Analysis with non-Gaussian Noise

Approximate Maximum Likelihood Estimations for Lognormal Distribution

Approximate Wireless Communication for Federated Learning

Maximum-Likelihood, Magnitude-Based, Amplitude and Noise Variance Estimation

Local-linear Likelihood Estimation of Geographically Weighted Generalised Linear Models