Carregant...
Carregant...

Vés al contingut (premeu Retorn)

Online EM with weight-based forgetting

Autor
Celaya, E.; Agostini, A.
Tipus d'activitat
Article en revista
Revista
Neural computation
Data de publicació
2015
Volum
27
Número
5
Pàgina inicial
1142
Pàgina final
1157
DOI
https://doi.org/10.1162/NECO_a_00723 Obrir en finestra nova
Repositori
http://hdl.handle.net/2117/81127 Obrir en finestra nova
Resum
In the on-line version of the EM algorithm introduced by Sato and Ishii (2000), a time-dependent discount factor is introduced for forgetting the effect of the old posterior values obtained with an earlier, inaccurate estimator. In their approach, forgetting is uniformly applied to the estimators of each mixture component depending exclusively on time, irrespective of the weight attributed to each unit for the observed sample. This causes an excessive forgetting in the less frequently sampled re...
Citació
Celaya, E., Agostini, A. Online EM with weight-based forgetting. "Neural computation", 2015, vol. 27, núm. 5, p. 1142-1157.
Paraules clau
learning (artificial intelligence), stochastic programming
Grup de recerca
KRD - Cinemàtica i Disseny de Robots

Participants