Deep Learning For Time Series Classification Using New Hand-Crafted Convolution Filters - Archive ouverte HAL Accéder directement au contenu
Communication Dans Un Congrès Année : 2022

Deep Learning For Time Series Classification Using New Hand-Crafted Convolution Filters

Résumé

In recent years, there has been an increasing interest in Deep Learning models for time series classification. In this field, state-of-the-art architectures rely on convolution neural networks that learn one dimensional filters in order to capture patterns allowing to discriminate between the different classes. These filters are randomly initialized and modified throughout model training. In this paper, we explore the creation of handcrafted (non learned) filters in order to capture specific patterns in a time series. We propose a set of filters whose values are fixed and not modified during the training step. Our goal with these filters is to capture specific patterns in a time series (increase, decrease, peaks) and study the relevance of adding such filters to existing architectures ranging from simple architecture (Fully Convolutional Network (FNC)) to state-of-the-art architecture (InceptionTime). Experiments reveal that adding our manually created filters increase the prediction accuracy on a majority of the 128 datasets of the UCR Archive. They also show that handcrafted filters and learned filters are complementary to obtain the best preforming models. This work is the first step in proposing a catalog of generic and fixed filters that could be useful in a large range of applications to improve deep models accuracy for time series classification.
Fichier principal
Vignette du fichier
bigdata2022.pdf (2.39 Mo) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)

Dates et versions

hal-04143093 , version 1 (27-06-2023)

Identifiants

Citer

Ali Ismail-Fawaz, Maxime Devanne, Jonathan Weber, Germain Forestier. Deep Learning For Time Series Classification Using New Hand-Crafted Convolution Filters. IEEE International Conference on Big Data (Big Data), Dec 2022, Osaka, Japan. pp.972-981, ⟨10.1109/bigdata55660.2022.10020496⟩. ⟨hal-04143093⟩
10 Consultations
19 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More