Speed-Constrained Tuning for Statistical Machine Translation Using Bayesian Optimization

TL;DR

This paper presents a Bayesian Optimization approach to tune statistical machine translation parameters, maximizing translation quality while satisfying decoding speed constraints efficiently across multiple language pairs.

Contribution

It introduces a method to incorporate speed constraints into Bayesian Optimization for SMT tuning, reducing optimization time and decoupling speed from BLEU measurement.

Findings

Significant reduction in optimization time compared to grid and random search.

Effective incorporation of speed constraints with confidence margins.

Decoupling speed measurement from BLEU evaluation further improves efficiency.

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.