Sciweavers

2 search results - page 1 / 1
» Softmax-Margin CRFs: Training Log-Linear Models with Cost Fu...
Sort
View
NAACL
2010
13 years 2 months ago
Softmax-Margin CRFs: Training Log-Linear Models with Cost Functions
We describe a method of incorporating taskspecific cost functions into standard conditional log-likelihood (CLL) training of linear structured prediction models. Recently introduc...
Kevin Gimpel, Noah A. Smith
EMNLP
2011
12 years 4 months ago
Training a Log-Linear Parser with Loss Functions via Softmax-Margin
Log-linear parsing models are often trained by optimizing likelihood, but we would prefer to optimise for a task-specific metric like Fmeasure. Softmax-margin is a convex objecti...
Michael Auli, Adam Lopez