| W09-1124 | conditional log-likelihood learning ( | L-BFGS optimization | ) . We used this latter method |
| P15-2028 | cross entropy error . We use the | L-BFGS optimization | algorithm to optimize our objective |
| P06-1028 | Sha and Pereira , 2003 ) with | L-BFGS optimization | . For MCE , we only considered |
| D13-1137 | parameters after t iterations of the | L-BFGS optimization | . Our preliminary experimental |
| D09-1014 | linear-chain CRF on the alignment - | L-BFGS optimization | procedure checks whether the |
| P14-1066 | Careful implementation of the | L-BFGS optimization | based on the BLEU - centric objective |
| N09-1007 | DPLVMs and CRFs . We apply the | L-BFGS optimization | algorithm to optimize the objective |
| D10-1061 | parator . We use the standard | L-BFGS optimization | algorithm ( Liu and Nocedal , |
| P14-2044 | Klementiev , 2012 ) . We interleave | L-BFGS optimization | within sampling , as in Monte |
| W05-1505 | the MaxEnt estimator using the | L-BFGS optimization | algorithms and Gaussian smoothing |
| W06-1643 | training log-linear models with | L-BFGS optimization | techniques and maximize the loglikelihood |
| P12-1109 | BILOU " encoding scheme . with | L-BFGS optimization | . We use the charac - ter/phoneme |
| P10-2028 | these five runs . Also we perform | L-BFGS optimization | to automatically adjust the non-informative |
| P08-4003 | of Liu and Nocedal 's ( 1989 ) | L-BFGS optimization | code , with a function for programmatic |
