15 search results - page 1 / 3 » A GIS-like training algorithm for log-linear models with hid... |
ICASSP
2008
IEEE
13 years 11 months ago
2008
IEEE
Conditional Random Fields (CRFs) are often estimated using an entropy based criterion in combination with Generalized Iterative Scaling (GIS). GIS offers, upon others, the immedi...
NIPS
2007
13 years 6 months ago
2007
We investigate a new, convex relaxation of an expectation-maximization (EM) variant that approximates a standard objective while eliminating local minima. First, a cautionary resu...
JMLR
2010
12 years 11 months ago
2010
We show how a preselection of hidden variables can be used to efficiently train generative models with binary hidden variables. The approach is based on Expectation Maximization (...
ICASSP
2011
IEEE
12 years 8 months ago
2011
IEEE
We have recently proposed an EM-style algorithm to optimize log-linear models with hidden variables. In this paper, we use this algorithm to optimize a hidden conditional random ï...
CSL
2007
Springer
2007
Springer
Speaker-adaptive learning of resonance targets in a hidden trajectory model of speech coarticulation
13 years 4 months ago
A novel speaker-adaptive learning algorithm is developed and evaluated for a hidden trajectory model of speech coarticulation and reduction. Central to this model is the process o...