| W07-0711 | Then we describe the equations of | MAP training | for word dependent transition |
| N06-1012 | the feature bags using standard | MAP training | , yielding individual CRFs { |
| W07-0711 | Gauvain and Lee , 1994 ) . In | MAP training | , an appropriate prior distribution |
| D14-1096 | approach arises from the prior : | MAP training | with a zero mean Gaussian prior |
| W07-0711 | , we can obtain the iterative | MAP training | formula for transition models |
| H92-1036 | densities was presented . Two | MAP training | algorithms , the forward-baclovard |
| P06-1028 | 2003 ) of 84.04 , which used the | MAP training | of CRFs with a feature selection |
