Speed-Constrained Tuning for Statistical Machine Translation Using Bayesian Optimization

Speed-Constrained Tuning for Statistical Machine Translation Using Bayesian Optimization ” by Daniel Beck, Adrià de Gispert, Gonzalo Iglesias, Aurelien Waite, and Bill Byrne. In Proceedings of the 2016 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies , 2016, pp. 856-865.

Abstract

We address the problem of automatically finding the parameters of a statistical machine translation system that maximize BLEU scores while ensuring that decoding speed exceeds a minimum value. We propose the use of Bayesian Optimization to efficiently tune the speed-related decoding parameters by easily incorporating speed as a noisy constraint function. The obtained parameter values are guaranteed to satisfy the speed constraint with an associated confidence margin. Across three language pairs and two speed constraint values, we report overall optimization time reduction compared to grid and random search. We also show that Bayesian Optimization can decouple speed and BLEU measurements, resulting in a further reduction of overall optimization time as speed is measured over a small subset of sentences.

BibTeX entry:

@inproceedings{N16-1100,
   author = {Daniel Beck and Adri{\`a} de Gispert and Gonzalo Iglesias and
	Aurelien Waite and Bill Byrne},
   title = {Speed-Constrained Tuning for Statistical Machine Translation
	Using Bayesian Optimization },
   booktitle = {Proceedings of the 2016 Conference of the North American
	Chapter of the Association for Computational Linguistics: Human
	Language Technologies },
   pages = {856--865},
   publisher = {Association for Computational Linguistics},
   year = {2016},
   url = {http://aclweb.org/anthology/N16-1100}
}

Back to Bill Byrne publications.