Decentralized Time Series Classification with ROCKET Features

Fuente: arXiv
Guardado en:
Detalles Bibliográficos
Autores principales: Casella, Bruno, Jakobs, Matthias, Aldinucci, Marco, Buschjäger, Sebastian
Formato: Preprint
Publicado: 2025
Materias:
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866909592267522048
author Casella, Bruno
Jakobs, Matthias
Aldinucci, Marco
Buschjäger, Sebastian
author_facet Casella, Bruno
Jakobs, Matthias
Aldinucci, Marco
Buschjäger, Sebastian
contents Time series classification (TSC) is a critical task with applications in various domains, including healthcare, finance, and industrial monitoring. Due to privacy concerns and data regulations, Federated Learning has emerged as a promising approach for learning from distributed time series data without centralizing raw information. However, most FL solutions rely on a client-server architecture, which introduces robustness and confidentiality risks related to the distinguished role of the server, which is a single point of failure and can observe knowledge extracted from clients. To address these challenges, we propose DROCKS, a fully decentralized FL framework for TSC that leverages ROCKET (RandOm Convolutional KErnel Transform) features. In DROCKS, the global model is trained by sequentially traversing a structured path across federation nodes, where each node refines the model and selects the most effective local kernels before passing them to the successor. Extensive experiments on the UCR archive demonstrate that DROCKS outperforms state-of-the-art client-server FL approaches while being more resilient to node failures and malicious attacks. Our code is available at https://anonymous.4open.science/r/DROCKS-7FF3/README.md.
format Preprint
id arxiv_https___arxiv_org_abs_2504_17617
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Decentralized Time Series Classification with ROCKET Features
Casella, Bruno
Jakobs, Matthias
Aldinucci, Marco
Buschjäger, Sebastian
Machine Learning
Artificial Intelligence
68T07
I.2.11; I.2.6
Time series classification (TSC) is a critical task with applications in various domains, including healthcare, finance, and industrial monitoring. Due to privacy concerns and data regulations, Federated Learning has emerged as a promising approach for learning from distributed time series data without centralizing raw information. However, most FL solutions rely on a client-server architecture, which introduces robustness and confidentiality risks related to the distinguished role of the server, which is a single point of failure and can observe knowledge extracted from clients. To address these challenges, we propose DROCKS, a fully decentralized FL framework for TSC that leverages ROCKET (RandOm Convolutional KErnel Transform) features. In DROCKS, the global model is trained by sequentially traversing a structured path across federation nodes, where each node refines the model and selects the most effective local kernels before passing them to the successor. Extensive experiments on the UCR archive demonstrate that DROCKS outperforms state-of-the-art client-server FL approaches while being more resilient to node failures and malicious attacks. Our code is available at https://anonymous.4open.science/r/DROCKS-7FF3/README.md.
title Decentralized Time Series Classification with ROCKET Features
topic Machine Learning
Artificial Intelligence
68T07
I.2.11; I.2.6
url https://arxiv.org/abs/2504.17617