| Paper: | SLP-L12.3 | ||
| Session: | Discriminative Training | ||
| Time: | Friday, May 19, 14:40 - 15:00 | ||
| Presentation: | Lecture | ||
| Topic: | Speech and Spoken Language Processing: Discriminative Training Methods | ||
| Title: | Training Algorithms for Hidden Conditional Random Fields | ||
| Authors: | Milind Mahajan, Asela Gunawardana, Alex Acero, Microsoft Research, United States | ||
| Abstract: | We investigate algorithms for training hidden conditional random fields (HCRFs) -- a class of direct models with hidden state sequences. We compare stochastic gradient ascent with the RProp algorithm, and investigate stochastic versions of RProp. We propose a new scheme for model flattening, and compare it to the state of the art. Finally we give experimental results on the TIMIT phone classification task showing how these training options interact, comparing HCRFs to HMMs trained using extended Baum-Welch as well as stochastic gradient methods. | ||
©2018 Conference Management Services, Inc. -||- email: webmaster@icassp2006.org -||- Last updated Friday, August 17, 2012