Federated Learning over a Wireless Network: Distributed User Selection through Random Access

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Sun, Chen, Ma, Shiyao, Zheng, Ce, Wu, Songtao, Cui, Tao, Lyu, Lingjuan
Natura: Preprint
Pubblicazione: 2023
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866916392991719424
author Sun, Chen
Ma, Shiyao
Zheng, Ce
Wu, Songtao
Cui, Tao
Lyu, Lingjuan
author_facet Sun, Chen
Ma, Shiyao
Zheng, Ce
Wu, Songtao
Cui, Tao
Lyu, Lingjuan
contents User selection has become crucial for decreasing the communication costs of federated learning (FL) over wireless networks. However, centralized user selection causes additional system complexity. This study proposes a network intrinsic approach of distributed user selection that leverages the radio resource competition mechanism in random access. Taking the carrier sensing multiple access (CSMA) mechanism as an example of random access, we manipulate the contention window (CW) size to prioritize certain users for obtaining radio resources in each round of training. Training data bias is used as a target scenario for FL with user selection. Prioritization is based on the distance between the newly trained local model and the global model of the previous round. To avoid excessive contribution by certain users, a counting mechanism is used to ensure fairness. Simulations with various datasets demonstrate that this method can rapidly achieve convergence similar to that of the centralized user selection approach.
format Preprint
id arxiv_https___arxiv_org_abs_2307_03758
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle Federated Learning over a Wireless Network: Distributed User Selection through Random Access
Sun, Chen
Ma, Shiyao
Zheng, Ce
Wu, Songtao
Cui, Tao
Lyu, Lingjuan
Machine Learning
Artificial Intelligence
Networking and Internet Architecture
User selection has become crucial for decreasing the communication costs of federated learning (FL) over wireless networks. However, centralized user selection causes additional system complexity. This study proposes a network intrinsic approach of distributed user selection that leverages the radio resource competition mechanism in random access. Taking the carrier sensing multiple access (CSMA) mechanism as an example of random access, we manipulate the contention window (CW) size to prioritize certain users for obtaining radio resources in each round of training. Training data bias is used as a target scenario for FL with user selection. Prioritization is based on the distance between the newly trained local model and the global model of the previous round. To avoid excessive contribution by certain users, a counting mechanism is used to ensure fairness. Simulations with various datasets demonstrate that this method can rapidly achieve convergence similar to that of the centralized user selection approach.
title Federated Learning over a Wireless Network: Distributed User Selection through Random Access
topic Machine Learning
Artificial Intelligence
Networking and Internet Architecture
url https://arxiv.org/abs/2307.03758