TRAINING ALGORITHMS FOR HIDDEN CONDITIONAL RANDOM FIELDS

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 flatttening, 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.

A discrminative models that generalize both HMM and CRF

sppech recognition and classification probelms.

the hidden state sequences and arbitrary dependencied upon the acoustic

This is the key difference between HCRFs and HMMs – HCRFs model the state sequence as being conditionally Markov given the observation sequence, while HMMs model the state sequence as being Markov, and each observation being independent of all others given the corresponding state.

如果HCRF真的是CRF的升级版的话，那么

与CRF原理上，实现上有什么不同之处？

比CRF好在哪里？跟HMM（比CRF）还弱的比，有意思么？

电子版，没看懂！！