arXiv subjects

Markus Freitag

Publications and source records attributed to Markus Freitag.

59 records · Page 4Linked to original sources

arXiv2017

Ensemble Distillation for Neural Machine Translation

Knowledge distillation describes a method for training a student network to perform better by learning from a stronger teacher network. Translating a sentence with an Neural Machine Translation (NMT) engine is time expensive and having a smaller model speeds up this process. We demonstrate how to transfer the translation quality of an ensemble and an oracle BLEU teacher network into a single NMT system. Further, we present translation improvements from a teacher network that has the same architecture and dimensions of the student network. As the training of the student model is still expensive, we introduce a data filtering method based on the knowledge of the teacher model that not only speeds up the training, but also leads to better translation quality. Our techniques need no code change and can be easily reproduced with any NMT architecture to speed up the decoding process.

cs.CL↗

arXiv2017

Beam Search Strategies for Neural Machine Translation

The basic concept in Neural Machine Translation (NMT) is to train a large Neural Network that maximizes the translation performance on a given parallel corpus. NMT is then using a simple left-to-right beam-search decoder to generate new translations that approximately maximize the trained conditional probability. The current beam search strategy generates the target sentence word by word from left-to- right while keeping a fixed amount of active candidates at each time step. First, this simple search is less adaptive as it also expands candidates whose scores are much worse than the current best. Secondly, it does not expand hypotheses if they are not within the best scoring candidates, even if their scores are close to the best one. The latter one can be avoided by increasing the beam size until no performance improvement can be observed. While you can reach better performance, this has the draw- back of a slower decoding speed. In this paper, we concentrate on speeding up the decoder by applying a more flexible beam search strategy whose candidate size may vary at each time step depending on the candidate scores. We speed up the original decoder by up to 43% for the two language pairs German-English and Chinese-English without losing any translation quality.

cs.CL↗

arXiv2017

Attention-based Vocabulary Selection for NMT Decoding

Neural Machine Translation (NMT) models usually use large target vocabulary sizes to capture most of the words in the target language. The vocabulary size is a big factor when decoding new sentences as the final softmax layer normalizes over all possible target words. To address this problem, it is widely common to restrict the target vocabulary with candidate lists based on the source sentence. Usually, the candidate lists are a combination of external word-to-word aligner, phrase table entries or most frequent words. In this work, we propose a simple and yet novel approach to learn candidate lists directly from the attention layer during NMT training. The candidate lists are highly optimized for the current NMT model and do not need any external computation of the candidate pool. We show significant decoding speedup compared with using the entire vocabulary, without losing any translation quality for two language pairs.

cs.CL↗

arXiv2017

Local System Voting Feature for Machine Translation System Combination

In this paper, we enhance the traditional confusion network system combination approach with an additional model trained by a neural network. This work is motivated by the fact that the commonly used binary system voting models only assign each input system a global weight which is responsible for the global impact of each input system on all translations. This prevents individual systems with low system weights from having influence on the system combination output, although in some situations this could be helpful. Further, words which have only been seen by one or few systems rarely have a chance of being present in the combined output. We train a local system voting model by a neural network which is based on the words themselves and the combinatorial occurrences of the different system outputs. This gives system combination the option to prefer other systems at different word positions even for the same sentence.

cs.CL↗

arXiv2016

Fast Domain Adaptation for Neural Machine Translation

Neural Machine Translation (NMT) is a new approach for automatic translation of text from one human language into another. The basic concept in NMT is to train a large Neural Network that maximizes the translation performance on a given parallel corpus. NMT is gaining popularity in the research community because it outperformed traditional SMT approaches in several translation tasks at WMT and other evaluation tasks/benchmarks at least for some language pairs. However, many of the enhancements in SMT over the years have not been incorporated into the NMT framework. In this paper, we focus on one such enhancement namely domain adaptation. We propose an approach for adapting a NMT system to a new domain. The main idea behind domain adaptation is that the availability of large out-of-domain training data and a small in-domain training data. We report significant gains with our proposed method in both automatic metrics and a human subjective evaluation metric on two language pairs. With our adaptation method, we show large improvement on the new domain while the performance of our general domain only degrades slightly. In addition, our approach is fast enough to adapt an already trained system to a new domain within few hours without the need to retrain the NMT model on the combined data which usually takes several days/weeks depending on the volume of the data.

cs.CL↗