A Discriminative Global Training Algorithm for Statistical MT

Proceedings of the 21st International Conference on Computational Linguistics and 44th Annual Meeting of the ACL, pages 721–728, Sydney, July 2006. c⃝2006 Association for Computational Linguistics A Discriminative Global Training Algorithm for Statistical MT Christoph Tillmann IBM T.J. Watson Research Center Yorktown Heights, N.Y. 10598 ctill@us.ibm.com Tong Zhang Yahoo! Research New York City, N.Y. 10011 tzhang@yahoo-inc.com Abstract This paper presents a novel training al- gorithm for a linearly-scored block se- quence translation model. The key com- ponent is a new procedure to directly op- timize the global scoring function used by a SMT decoder. No translation, language, or distortion model probabilities are used as in earlier work on SMT. Therefore our method, which employs less domain specific knowledge, is both simpler and more extensible than previous approaches. Moreover, the training procedure treats the decoder as a black-box, and thus can be used to optimize any decoding scheme. The training algorithm is evaluated on a standard Arabic-English translation task. 1 Introduction This paper presents a view of phrase-based SMT as a sequential process that generates block ori- entation sequences. A block is a pair of phrases which are translations of each other. For example, Figure 1 shows an Arabic-English translation ex- ample that uses four blocks. During decoding, we view translation as a block segmentation process, where the input sentence is segmented from left to right and the target sentence is generated from bottom to top, one block at a time. A monotone block sequence is generated except for the possi- bility to handle some local phrase re-ordering. In this local re-ordering model (Tillmann and Zhang, 2005; Kumar and Byrne, 2005) a block � with orientation � is generated relative to its predeces- sor block �� . During decoding, we maximize the score ��� � � � �  of a block orientation sequence   ! "  # $�% !  &�$� !' ( ! ) (+* $ ( " * ( ! , $ - . ( ! (  $ ( - ! - . "  "/ 0  $�%1 ! !   (  - ( ! 2+3 ( ! ( !3 # - ( 4 5 6 Figure 1: An Arabic-English block translation ex- ample, where the Arabic words are romanized. The following orientation sequence is generated: � 87:9 ��; 7=< �?> 7:9 ��@ 7=A . � �B� � � ): � � � � � � �  7 � CED FHGJIK � � C � C L� CEM N (1) where � C is a block, � CEM is its predecessor block, and � CPORQ < � eft N A � ight N 9 � eutral NS is a three- valued orientation component linked to the block � C : a block is generated to the left or the right of its predecessor block � CTM , where the orientation � CEM of the predecessor block is ignored. Here, is the number of blocks in the translation. We areU interested in …

Content truncated to 500 words