Federated Learning with MMD-based Early Stopping for Adaptive GNSS Interference Classification

Federated learning (FL) enables multiple devices to collaboratively train a global model while maintaining data on local servers. Each device trains the model on its local server and shares only the model updates (i.e., gradient weights) during the aggregation step. A significant challenge in FL is...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Gaikwad, Nishant S, Heublein, Lucas, Raichur, Nisha L, Feigl, Tobias, Mutschler, Christopher, Ott, Felix
Format: Artikel
Sprache:eng
Schlagworte:
Online-Zugang:Volltext bestellen
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
container_end_page
container_issue
container_start_page
container_title
container_volume
creator Gaikwad, Nishant S
Heublein, Lucas
Raichur, Nisha L
Feigl, Tobias
Mutschler, Christopher
Ott, Felix
description Federated learning (FL) enables multiple devices to collaboratively train a global model while maintaining data on local servers. Each device trains the model on its local server and shares only the model updates (i.e., gradient weights) during the aggregation step. A significant challenge in FL is managing the feature distribution of novel, unbalanced data across devices. In this paper, we propose an FL approach using few-shot learning and aggregation of the model weights on a global server. We introduce a dynamic early stopping method to balance out-of-distribution classes based on representation learning, specifically utilizing the maximum mean discrepancy of feature embeddings between local and global models. An exemplary application of FL is orchestrating machine learning models along highways for interference classification based on snapshots from global navigation satellite system (GNSS) receivers. Extensive experiments on four GNSS datasets from two real-world highways and controlled environments demonstrate that our FL method surpasses state-of-the-art techniques in adapting to both novel interference classes and multipath scenarios.
doi_str_mv 10.48550/arxiv.2410.15681
format Article
fullrecord <record><control><sourceid>arxiv_GOX</sourceid><recordid>TN_cdi_arxiv_primary_2410_15681</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>2410_15681</sourcerecordid><originalsourceid>FETCH-arxiv_primary_2410_156813</originalsourceid><addsrcrecordid>eNqFjrEOgjAURbs4GPUDnOwPgKBgWA2CmqgLzpInvGoTLM1rg_L3AnF3usm5ZziMzX3PDaIw9JZAH9m4q6ADfriJ_DG7pVgigcWSnxBISfXgb2mf_HzeOXcwHU-AqpZntta6f0VNfFuCtrJBvr9kGT8qiySQUBXI4wqMkUIWYGWtpmwkoDI4--2ELdLkGh-coSTXJF9Abd4X5UPR-r_xBWz4QSc</addsrcrecordid><sourcetype>Open Access Repository</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype></control><display><type>article</type><title>Federated Learning with MMD-based Early Stopping for Adaptive GNSS Interference Classification</title><source>arXiv.org</source><creator>Gaikwad, Nishant S ; Heublein, Lucas ; Raichur, Nisha L ; Feigl, Tobias ; Mutschler, Christopher ; Ott, Felix</creator><creatorcontrib>Gaikwad, Nishant S ; Heublein, Lucas ; Raichur, Nisha L ; Feigl, Tobias ; Mutschler, Christopher ; Ott, Felix</creatorcontrib><description>Federated learning (FL) enables multiple devices to collaboratively train a global model while maintaining data on local servers. Each device trains the model on its local server and shares only the model updates (i.e., gradient weights) during the aggregation step. A significant challenge in FL is managing the feature distribution of novel, unbalanced data across devices. In this paper, we propose an FL approach using few-shot learning and aggregation of the model weights on a global server. We introduce a dynamic early stopping method to balance out-of-distribution classes based on representation learning, specifically utilizing the maximum mean discrepancy of feature embeddings between local and global models. An exemplary application of FL is orchestrating machine learning models along highways for interference classification based on snapshots from global navigation satellite system (GNSS) receivers. Extensive experiments on four GNSS datasets from two real-world highways and controlled environments demonstrate that our FL method surpasses state-of-the-art techniques in adapting to both novel interference classes and multipath scenarios.</description><identifier>DOI: 10.48550/arxiv.2410.15681</identifier><language>eng</language><subject>Computer Science - Distributed, Parallel, and Cluster Computing ; Computer Science - Learning</subject><creationdate>2024-10</creationdate><rights>http://arxiv.org/licenses/nonexclusive-distrib/1.0</rights><oa>free_for_read</oa><woscitedreferencessubscribed>false</woscitedreferencessubscribed></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><link.rule.ids>228,230,780,885</link.rule.ids><linktorsrc>$$Uhttps://arxiv.org/abs/2410.15681$$EView_record_in_Cornell_University$$FView_record_in_$$GCornell_University$$Hfree_for_read</linktorsrc><backlink>$$Uhttps://doi.org/10.48550/arXiv.2410.15681$$DView paper in arXiv$$Hfree_for_read</backlink></links><search><creatorcontrib>Gaikwad, Nishant S</creatorcontrib><creatorcontrib>Heublein, Lucas</creatorcontrib><creatorcontrib>Raichur, Nisha L</creatorcontrib><creatorcontrib>Feigl, Tobias</creatorcontrib><creatorcontrib>Mutschler, Christopher</creatorcontrib><creatorcontrib>Ott, Felix</creatorcontrib><title>Federated Learning with MMD-based Early Stopping for Adaptive GNSS Interference Classification</title><description>Federated learning (FL) enables multiple devices to collaboratively train a global model while maintaining data on local servers. Each device trains the model on its local server and shares only the model updates (i.e., gradient weights) during the aggregation step. A significant challenge in FL is managing the feature distribution of novel, unbalanced data across devices. In this paper, we propose an FL approach using few-shot learning and aggregation of the model weights on a global server. We introduce a dynamic early stopping method to balance out-of-distribution classes based on representation learning, specifically utilizing the maximum mean discrepancy of feature embeddings between local and global models. An exemplary application of FL is orchestrating machine learning models along highways for interference classification based on snapshots from global navigation satellite system (GNSS) receivers. Extensive experiments on four GNSS datasets from two real-world highways and controlled environments demonstrate that our FL method surpasses state-of-the-art techniques in adapting to both novel interference classes and multipath scenarios.</description><subject>Computer Science - Distributed, Parallel, and Cluster Computing</subject><subject>Computer Science - Learning</subject><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2024</creationdate><recordtype>article</recordtype><sourceid>GOX</sourceid><recordid>eNqFjrEOgjAURbs4GPUDnOwPgKBgWA2CmqgLzpInvGoTLM1rg_L3AnF3usm5ZziMzX3PDaIw9JZAH9m4q6ADfriJ_DG7pVgigcWSnxBISfXgb2mf_HzeOXcwHU-AqpZntta6f0VNfFuCtrJBvr9kGT8qiySQUBXI4wqMkUIWYGWtpmwkoDI4--2ELdLkGh-coSTXJF9Abd4X5UPR-r_xBWz4QSc</recordid><startdate>20241021</startdate><enddate>20241021</enddate><creator>Gaikwad, Nishant S</creator><creator>Heublein, Lucas</creator><creator>Raichur, Nisha L</creator><creator>Feigl, Tobias</creator><creator>Mutschler, Christopher</creator><creator>Ott, Felix</creator><scope>AKY</scope><scope>GOX</scope></search><sort><creationdate>20241021</creationdate><title>Federated Learning with MMD-based Early Stopping for Adaptive GNSS Interference Classification</title><author>Gaikwad, Nishant S ; Heublein, Lucas ; Raichur, Nisha L ; Feigl, Tobias ; Mutschler, Christopher ; Ott, Felix</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-arxiv_primary_2410_156813</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2024</creationdate><topic>Computer Science - Distributed, Parallel, and Cluster Computing</topic><topic>Computer Science - Learning</topic><toplevel>online_resources</toplevel><creatorcontrib>Gaikwad, Nishant S</creatorcontrib><creatorcontrib>Heublein, Lucas</creatorcontrib><creatorcontrib>Raichur, Nisha L</creatorcontrib><creatorcontrib>Feigl, Tobias</creatorcontrib><creatorcontrib>Mutschler, Christopher</creatorcontrib><creatorcontrib>Ott, Felix</creatorcontrib><collection>arXiv Computer Science</collection><collection>arXiv.org</collection></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext_linktorsrc</fulltext></delivery><addata><au>Gaikwad, Nishant S</au><au>Heublein, Lucas</au><au>Raichur, Nisha L</au><au>Feigl, Tobias</au><au>Mutschler, Christopher</au><au>Ott, Felix</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>Federated Learning with MMD-based Early Stopping for Adaptive GNSS Interference Classification</atitle><date>2024-10-21</date><risdate>2024</risdate><abstract>Federated learning (FL) enables multiple devices to collaboratively train a global model while maintaining data on local servers. Each device trains the model on its local server and shares only the model updates (i.e., gradient weights) during the aggregation step. A significant challenge in FL is managing the feature distribution of novel, unbalanced data across devices. In this paper, we propose an FL approach using few-shot learning and aggregation of the model weights on a global server. We introduce a dynamic early stopping method to balance out-of-distribution classes based on representation learning, specifically utilizing the maximum mean discrepancy of feature embeddings between local and global models. An exemplary application of FL is orchestrating machine learning models along highways for interference classification based on snapshots from global navigation satellite system (GNSS) receivers. Extensive experiments on four GNSS datasets from two real-world highways and controlled environments demonstrate that our FL method surpasses state-of-the-art techniques in adapting to both novel interference classes and multipath scenarios.</abstract><doi>10.48550/arxiv.2410.15681</doi><oa>free_for_read</oa></addata></record>
fulltext fulltext_linktorsrc
identifier DOI: 10.48550/arxiv.2410.15681
ispartof
issn
language eng
recordid cdi_arxiv_primary_2410_15681
source arXiv.org
subjects Computer Science - Distributed, Parallel, and Cluster Computing
Computer Science - Learning
title Federated Learning with MMD-based Early Stopping for Adaptive GNSS Interference Classification
url https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2024-12-23T20%3A11%3A28IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-arxiv_GOX&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.genre=article&rft.atitle=Federated%20Learning%20with%20MMD-based%20Early%20Stopping%20for%20Adaptive%20GNSS%20Interference%20Classification&rft.au=Gaikwad,%20Nishant%20S&rft.date=2024-10-21&rft_id=info:doi/10.48550/arxiv.2410.15681&rft_dat=%3Carxiv_GOX%3E2410_15681%3C/arxiv_GOX%3E%3Curl%3E%3C/url%3E&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_id=info:pmid/&rfr_iscdi=true