Pré-Publication, Document De Travail Année : 2024

Time series classification with random convolution kernels: pooling operators and input representations matter

Résumé

This article presents a new approach based on MiniRocket, called SelF-Rocket, for fast time series classification (TSC). Unlike existing approaches based on random convolution kernels, it dynamically selects the best couple of input representations and pooling operator during the training process. SelF-Rocket achieves state-of-the-art accuracy on the University of California Riverside (UCR) TSC benchmark datasets.

Fichier non déposé

Dates et versions

hal-05162026 , version 1 (15-07-2025)

Identifiants

Citer

Mouhamadou Mansour Lo, Gildas Morvan, Mathieu Rossi, Fabrice Morganti, David Mercier. Time series classification with random convolution kernels: pooling operators and input representations matter. 2025. ⟨hal-05162026⟩
45 Consultations
0 Téléchargements

Altmetric

Partager

  • More