Sciweavers

Share
EMNLP
2009

Consensus Training for Consensus Decoding in Machine Translation

10 years 11 months ago
Consensus Training for Consensus Decoding in Machine Translation
We propose a novel objective function for discriminatively tuning log-linear machine translation models. Our objective explicitly optimizes the BLEU score of expected n-gram counts, the same quantities that arise in forestbased consensus and minimum Bayes risk decoding methods. Our continuous objective can be optimized using simple gradient ascent. However, computing critical quantities in the gradient necessitates a novel dynamic program, which we also present here. Assuming BLEU as an evaluation measure, our objective function has two principle advantages over standard max BLEU tuning. First, it specifically optimizes model weights for downstream consensus decoding procedures. An unexpected second benefit is that it reduces overfitting, which can improve test set BLEU scores when using standard Viterbi decoding.
Adam Pauls, John DeNero, Dan Klein
Added 17 Feb 2011
Updated 17 Feb 2011
Type Journal
Year 2009
Where EMNLP
Authors Adam Pauls, John DeNero, Dan Klein
Comments (0)
books