# LATTICE-BASED MINIMUM ERROR RATE TRAINING USING WEIGHTED FINITE-STATE TRANSDUCERS WITH TROPICAL POLYNOMIAL WEIGHTS

LATTICE-BASED MINIMUM ERROR RATE TRAINING USING WEIGHTED FINITE-STATE TRANSDUCERS WITH TROPICAL POLYNOMIAL WEIGHTS” by A. Waite, G. Blackwood, and W. Byrne. In Proceedings of the 10^th International Workshop on Finite State Methods and Natural Language Processing (FSMNLP 2012), (Donostia-San Sebastian, Spain), July 2012.

## Abstract

Minimum Error Rate Training (MERT) is a method for training the parameters of a log-linear model. One advantage of this method of training is that it can use the large number of hypotheses encoded in a translation lattice as training data. We demonstrate that the MERT line optimisation can be modelled as computing the shortest distance in a weighted finite-state transducer using a tropical polynomial semiring.

BibTeX entry:

@inproceedings{fsmnlp12,
author = {A. Waite and G. Blackwood and W. Byrne},
title = {LATTICE-BASED MINIMUM ERROR RATE TRAINING USING WEIGHTED
FINITE-STATE TRANSDUCERS WITH TROPICAL POLYNOMIAL WEIGHTS},
booktitle = {Proceedings of the 10{\it ^{th}} International Workshop on
Finite State Methods and Natural Language Processing (FSMNLP
2012)},
pages = {(11 pages)},