To access the full text documents, please follow this link: http://hdl.handle.net/2117/15675

An adaptive gradient-search based algorithm for discriminative training of hmm's
Nogueiras Rodríguez, Albino; Mariño Acebal, José Bernardo; Monte Moreno, Enrique
Universitat Politècnica de Catalunya. Departament de Teoria del Senyal i Comunicacions; Universitat Politècnica de Catalunya. VEU - Grup de Tractament de la Parla
Although having revealed to be a very powerful tool in acoustic modelling, discriminative training presents a major drawback: the lack of a formulation guaranteeing convergence in no matter which initial conditions, such as the Baum-Welch algorithm in maximum likelihood training. For this reason, a gradient descent search is usually used in this kind of problem. Unfortunately, standard gradient descent algorithms rely heavily on the election of the learning rates. This dependence is specially cumbersome because it represents that, at each run of the discriminative training procedure, a search should be carried out over the parameters ruling the algorithm. In this paper we describe an adaptive procedure for determining the optimal value of the step size at each iteration. While the calculus and memory overhead of the algorithm is negligible, results show less dependence on the initial learning rate than standard gradient descent and, using the same idea in order to apply self-scaling, it clearly outperforms it.
Peer Reviewed
-Àrees temàtiques de la UPC::Informàtica::Intel·ligència artificial::Llenguatge natural
-Natural language processing (Computer science)
-Processament en llenguatge natural (Informàtica)
Article - Published version
Conference Object
Robert H. Mannel and Jordi Robert-Ribes
         

Show full item record

Related documents

Other documents of the same author

Nogueiras Rodríguez, Albino; Caballero Galeote, Mónica; Mariño Acebal, José Bernardo
Mariño Acebal, José Bernardo; Nogueiras Rodríguez, Albino; Pachés-Leal, Pau; Bonafonte Cávez, Antonio
Bonafonte Cávez, Antonio; Mariño Acebal, José Bernardo; Nogueiras Rodríguez, Albino
Nadeu Camprubí, Climent; Mariño Acebal, José Bernardo; Hernando Pericás, Francisco Javier; Nogueiras Rodríguez, Albino
 

Coordination

 

Supporters