## Efficient Dynamic Programming Search Algorithms for Phrase-Based SMT

### Cached

### Download Links

### BibTeX

@MISC{_efficientdynamic,

author = {},

title = {Efficient Dynamic Programming Search Algorithms for Phrase-Based SMT},

year = {}

}

### OpenURL

### Abstract

This paper presents a series of efficient dynamic-programming (DP) based algorithms for phrase-based decoding and alignment computation in statistical machine translation (SMT). The DP-based decoding algorithms are analyzed in terms of shortest path-finding algorithms, where the similarity to DP-based decoding algorithms in speech recognition is demonstrated. The paper contains the following original contributions: 1) the DP-based decoding algorithm in (Tillmann and Ney, 2003) is extended in a formal way to handle phrases and a novel pruning strategy with increased translation speed is presented 2) a novel alignment algorithm is presented that computes a phrase alignment efficiently in the case that it is consistent with an underlying word alignment. Under certain restrictions, both algorithms handle MT-related problems efficiently that are generally NP complete (Knight, 1999). 1

### Citations

797 |
Statistical methods for speech recognition
- Jelinek
- 1997
(Show Context)
Citation Context ... Introduction This paper deals with dynamic programming based decoding and alignment algorithms for phrase-based SMT. Dynamic Programming based search algorithms are being used in speech recognition (=-=Jelinek, 1998-=-; Ney et al., 1992) as well as in statistical machine translation (Tillmann et al., 1997; Niessen et al., 1998; Tillmann and Ney, 2003). Here, the decoding algorithms are described as shortest path fi... |

400 |
The alignment template approach to statistical machine translation
- Och, Ney
- 2004
(Show Context)
Citation Context ...estriction is demonstrated in the following section. 4 Efficient Block Alignment Algorithm A common approach to phrase-based SMT is to learn phrasal translation pairs from word-aligned training data (=-=Och and Ney, 2004-=-). Here, a word alignment � is a subset of the Cartesian product of source and target positions: � ����� �§� ������� � ����� ��� ��� ������� � Here, is the target sentence length � and is the source s... |

140 |
A dynamic programming approach to sequencing problems
- Held, Karp
- 1962
(Show Context)
Citation Context ...gorithm can be described as a shortest path finding algorithm. But this has been shown by linking decoding to a dynamic-programming solution for the traveling salesman problem. This algorithm due to (=-=Held and Karp, 1962-=-) is a special case of a shortest path finding algorithm (Dreyfus and Law, 1977). The regularly structured search graph for this problem is illustrated in Fig. 1: all paths from the left-most to the r... |

129 | Decoding complexity in wordreplacement translation models
- Knight
- 1999
(Show Context)
Citation Context ...ent efficiently in the case that it is consistent with an underlying word alignment. Under certain restrictions, both algorithms handle MT-related problems efficiently that are generally NP complete (=-=Knight, 1999-=-). 1 Introduction This paper deals with dynamic programming based decoding and alignment algorithms for phrase-based SMT. Dynamic Programming based search algorithms are being used in speech recogniti... |

81 | Introduction to Algorithms - Rivest - 1990 |

77 |
The art and theory of dynamic programming
- Dreyfus, Law
- 1977
(Show Context)
Citation Context ...een shown by linking decoding to a dynamic-programming solution for the traveling salesman problem. This algorithm due to (Held and Karp, 1962) is a special case of a shortest path finding algorithm (=-=Dreyfus and Law, 1977-=-). The regularly structured search graph for this problem is illustrated in Fig. 1: all paths from the left-most to the right-most vertex correspond to a translation of the inFinalsput sentence, where... |

67 |
The harpy speech understanding system
- Lowerre, Reddy
- 1980
(Show Context)
Citation Context ...� � � ��� � ��� � ��� (3) The ��� state fields are updated on a component-bycomponent basis. ������������� � ����� is the coverage vec11 Table 2: Single-beam (� -Beam) decoding algorithm (related to (=-=Lowerre and Reddy, 1980-=-)). input: source sentence with words � � � ����� � ��� ������������� for each � � � � ����� ��� do ��� ������������� for each � state � in do CLOSED?¤§��� if then matcher: for ��������������� each el... |

66 |
Language translation apparatus and method of using context-based translation models
- Berger, Brown, et al.
- 1996
(Show Context)
Citation Context ... 1 The multi-beam decoder is similar to the decoder presented in (Koehn, 2004) which is a standard decoder used in phrasebased SMT. A multi-beam decoder is also used in (Al-Onaizan et al., 2004) and (=-=Berger et al., 1996-=-). 2 In that work, there is a distinction between within-word and between-word search, which is not relevant for phrase-based decoding where only exact phrase matches are searched. 10 � � ���������£� ... |

46 | A DP based Search Using Monotone Alignments in Statistical Translation
- Tillmann, Vogel, et al.
- 1997
(Show Context)
Citation Context ...ment algorithms for phrase-based SMT. Dynamic Programming based search algorithms are being used in speech recognition (Jelinek, 1998; Ney et al., 1992) as well as in statistical machine translation (=-=Tillmann et al., 1997-=-; Niessen et al., 1998; Tillmann and Ney, 2003). Here, the decoding algorithms are described as shortest path finding algorithms in regularly structured search graphs or search grids. Under certain re... |

42 |
Projection Extension Algorithm for Statistical Machine Translation
- Tillmann, “A
(Show Context)
Citation Context ...ining data efficiently. To restrict the block selection based on word aligned training data, interval projection functions are defined as follows 4 � : is a source interval � and is an target inter3 (=-=Tillmann, 2003-=-) reports an intersection coverage of about % for Arabic-English parallel data, and a coverage of ��� ��� % for Chinese-English data. In the case of uncomplete coverage, the current algorithm can be e... |

33 | A DP based Search Algorithm for Statistical Machine Translation
- Nießen, Vogel, et al.
- 1998
(Show Context)
Citation Context ...ase-based SMT. Dynamic Programming based search algorithms are being used in speech recognition (Jelinek, 1998; Ney et al., 1992) as well as in statistical machine translation (Tillmann et al., 1997; =-=Niessen et al., 1998-=-; Tillmann and Ney, 2003). Here, the decoding algorithms are described as shortest path finding algorithms in regularly structured search graphs or search grids. Under certain restrictions, e.g. start... |