MAD: Match-And-Deform for Time Series Domain Adaptation - Archive ouverte HAL Accéder directement au contenu
Communication Dans Un Congrès Année : 2022

MAD: Match-And-Deform for Time Series Domain Adaptation

Résumé

While large volumes of unlabeled data are usually available, associated labels are often scarce. The unsupervised Domain Adaptation problem aims at exploiting labels from a source domain to classify data from a related, yet different, target domain. When time series are at stake, new difficulties arise as temporal shifts may appear in addition to the standard feature distribution shift. In this paper, we introduce the Match-And-Deform (MAD) approach that aims at finding correspondences between the source and target time series while taking into account the temporal distortions that may occur. The associated optimization problem allows simultaneously aligning the series thanks to an optimal transport loss and the time stamps through dynamic time warping. When embedded into a deep neural network, MAD helps learning new representations of time series that both align the domains and maximizes the discriminative power of the network. Empirical studies on benchmark datasets and remote sensing data demonstrate that MAD makes meaningful sample-to-sample pairing and time shift estimation, reaching similar or better classification performance than state-of-the-art deep time series domain adaptation strategies.
Fichier non déposé

Dates et versions

hal-03932463 , version 1 (10-01-2023)

Identifiants

  • HAL Id : hal-03932463 , version 1

Citer

François Painblanc, Laetitia Chapel, Nicolas Courty, Chloé Friguet, Charlotte Pelletier, et al.. MAD: Match-And-Deform for Time Series Domain Adaptation. Conférence sur l'Apprentissage automatique (CAp), Jul 2022, Vannes, France. ⟨hal-03932463⟩
63 Consultations
0 Téléchargements

Partager

Gmail Facebook X LinkedIn More