Time Window Width Influence On Dynamic BPTT(h) Learning Algorithm Performances: Experimental Study - Archive ouverte HAL Accéder directement au contenu
Communication Dans Un Congrès Année : 2006

Time Window Width Influence On Dynamic BPTT(h) Learning Algorithm Performances: Experimental Study

Résumé

The purpose of the research addressed in this paper is to study the influence of the time window width in dynamic truncated BackPropagation Through Time BPTT(h) learning algorithms. Statistical experiments based on the identification of a real biped robot balancing mechanism are carried out to raise the link between the window width and the stability, the speed and the accuracy of the learning. The time window width choice is shown to be crucial for the convergence speed of the learning process and the generalization ability of the network. Although, a particular attention is brought to a divergence problem (gradient blow up) observed with the assumption where the net parameters are constant along the window. The limit of this assumption is demonstrated and parameters evolution storage, used as a solution for this problem, is detailed.
Fichier principal
Vignette du fichier
ICANN06_HAL.pdf (855.77 Ko) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)

Dates et versions

hal-00523037 , version 1 (08-11-2022)

Identifiants

Citer

Vincent Scesa, Patrick Henaff, Fethi Ben Ouezdou, Faycal Namoun. Time Window Width Influence On Dynamic BPTT(h) Learning Algorithm Performances: Experimental Study. 16th International Conference on Artificial Neural Networks, ICANN 2006, Sep 2006, athene, Greece. pp.93 -102, ⟨10.1007/11840817_10⟩. ⟨hal-00523037⟩

Collections

UVSQ LISV
46 Consultations
21 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More