Dynamic factor mixture of experts for functional time series modeling - Archive ouverte HAL Accéder directement au contenu
Communication Dans Un Congrès Année : 2016

Dynamic factor mixture of experts for functional time series modeling

Résumé

A new approach is introduced in this paper for dynamic modeling and dimensionality reduction from time series of curves. For this purpose, a dynamic mixture of experts model whose regression coefficients evolve from curve to curve according to a Gaussian random walk over low dimensional factors, is proposed. The resulting model is neither else than a particular state-space model involving discrete and continuous latent variables, whose parameters are learned across a sequence of curves through a dedicated variational Expectation-Maximization algorithm. The experimental study conducted on simulated sequences of curves has shown the strong potential of the proposed approach.
Fichier non déposé

Dates et versions

hal-01395237 , version 1 (10-11-2016)

Identifiants

  • HAL Id : hal-01395237 , version 1

Citer

Allou Same. Dynamic factor mixture of experts for functional time series modeling. IEEE ICMLA'16 - IEEE International Conference on Machine Learning and Applications, Dec 2016, Anaheim, United States. 7p. ⟨hal-01395237⟩
44 Consultations
0 Téléchargements

Partager

Gmail Facebook X LinkedIn More