PyTorchKR Papers

최신 논문

총 5,206편
태그·기간 등 상세 필터

카테고리 탐색

로그인 후 이 태그 구독

arXiv:1709.00831

Hypothesis Testing based Intrinsic Evaluation of Word Embeddings

We introduce the cross-match test - an exact, distribution free, high-dimensional hypothesis test as an intrinsic evaluation metric for word embeddings. We show that cross-match is an effective means of measuring distributional similarity between different vector representations and of evaluating the statistical significance of different vector embedding models. Additionally, we find that cross-match can be used to provide a quantitative measure of linguistic similarity for selecting bridge languages for machine translation. We demonstrate that the results of the hypothesis test align with our expectations and note that the framework of two sample hypothesis testing is not limited to word embeddings and can be extended to all vector representations.

arXiv:1709.00616

Challenging Language-Dependent Segmentation for Arabic: An Application to Machine Translation and Part-of-Speech Tagging

Word segmentation plays a pivotal role in improving any Arabic NLP application. Therefore, a lot of research has been spent in improving its accuracy. Off-the-shelf tools, however, are: i) complicated to use and ii) domain/dialect dependent. We explore three language-independent alternatives to morphological segmentation using: i) data-driven sub-word units, ii) characters as a unit of learning, and iii) word embeddings learned using a character CNN (Convolution Neural Network). On the tasks of Machine Translation and POS tagging, we found these methods to achieve close to, and occasionally surpass state-of-the-art performance. In our analysis, we show that a neural machine translation system is sensitive to the ratio of source and target tokens, and a ratio close to 1 or greater, gives optimal performance.

arXiv:1708.08712

Neural Machine Translation Training in a Multi-Domain Scenario

In this paper, we explore alternative ways to train a neural machine translation system in a multi-domain scenario. We investigate data concatenation (with fine tuning), model stacking (multi-level fine tuning), data selection and multi-model ensemble. Our findings show that the best translation quality can be achieved by building an initial system on a concatenation of available out-of-domain data and then fine-tuning it on in-domain data. Model stacking works best when training begins with the furthest out-of-domain data and the model is incrementally fine-tuned with the next furthest domain and so on. Data selection did not give the best results, but can be considered as a decent compromise between training time and translation quality. A weighted ensemble of different individual models performed better than data selection. It is beneficial in a scenario when there is no time for fine-tuning an already trained model.

arXiv:1707.06012

Modeling Target-Side Inflection in Neural Machine Translation

NMT systems have problems with large vocabulary sizes. Byte-pair encoding (BPE) is a popular approach to solving this problem, but while BPE allows the system to generate any target-side word, it does not enable effective generalization over the rich vocabulary in morphologically rich languages with strong inflectional phenomena. We introduce a simple approach to overcome this problem by training a system to produce the lemma of a word and its morphologically rich POS tag, which is then followed by a deterministic generation step. We apply this strategy for English-Czech and English-German translation scenarios, obtaining improvements in both settings. We furthermore show that the improvement is not due to only adding explicit morphological information.

arXiv:1704.04520

Neural Machine Translation Model with a Large Vocabulary Selected by Branching Entropy

Neural machine translation (NMT), a new approach to machine translation, has achieved promising results comparable to those of traditional approaches such as statistical machine translation (SMT). Despite its recent success, NMT cannot handle a larger vocabulary because the training complexity and decoding complexity proportionally increase with the number of target words. This problem becomes even more serious when translating patent documents, which contain many technical terms that are observed infrequently. In this paper, we propose to select phrases that contain out-of-vocabulary words using the statistical approach of branching entropy. This allows the proposed NMT system to be applied to a translation task of any language pair without any language-specific knowledge about technical term identification. The selected phrases are then replaced with tokens during training and post-translated by the phrase translation table of SMT. Evaluation on Japanese-to-Chinese, Chinese-to-Japanese, Japanese-to-English and English-to-Japanese patent sentence translation proved the effectiveness of phrases selected with branching entropy, where the proposed NMT model achieves a substantial improvement over a baseline NMT model without our proposed technique. Moreover, the number of translation errors of under-translation by the baseline NMT model without our proposed technique reduces to around half by the proposed NMT model.

arXiv:1709.02184

Translating Terminological Expressions in Knowledge Bases with Neural Machine Translation

Our work presented in this paper focuses on the translation of terminological expressions represented in semantically structured resources, like ontologies or knowledge graphs. The challenge of translating ontology labels or terminological expressions documented in knowledge bases lies in the highly specific vocabulary and the lack of contextual information, which can guide a machine translation system to translate ambiguous words into the targeted domain. Due to these challenges, we evaluate the translation quality of domain-specific expressions in the medical and financial domain with statistical as well as with neural machine translation methods and experiment domain adaptation of the translation models with terminological expressions only. Furthermore, we perform experiments on the injection of external terminological expressions into the translation systems. Through these experiments, we observed a significant advantage in domain adaptation for the domain-specific resource in the medical and financial domain and the benefit of subword models over word-based neural machine translation models for terminology translation.

arXiv:1709.01915

Towards Neural Machine Translation with Latent Tree Attention

Building models that take advantage of the hierarchical structure of language without a priori annotation is a longstanding goal in natural language processing. We introduce such a model for the task of machine translation, pairing a recurrent neural network grammar encoder with a novel attentional RNNG decoder and applying policy gradient reinforcement learning to induce unsupervised tree structures on both the source and target. When trained on character-level datasets with no explicit segmentation or parse annotation, the model learns a plausible segmentation and shallow parse, obtaining performance close to an attentional baseline.

arXiv:1709.03814

SYSTRAN Purely Neural MT Engines for WMT2017

This paper describes SYSTRAN's systems submitted to the WMT 2017 shared news translation task for English-German, in both translation directions. Our systems are built using OpenNMT, an open-source neural machine translation system, implementing sequence-to-sequence models with LSTM encoder/decoders and attention. We experimented using monolingual data automatically back-translated. Our resulting models are further hyper-specialised with an adaptation technique that finely tunes models according to the evaluation test sentences.

arXiv:1612.06140

Domain Control for Neural Machine Translation

Machine translation systems are very sensitive to the domains they were trained on. Several domain adaptation techniques have been deeply studied. We propose a new technique for neural machine translation (NMT) that we call domain control which is performed at runtime using a unique neural network covering multiple domains. The presented approach shows quality improvements when compared to dedicated domains translating on any of the covered domains and even on out-of-domain data. In addition, model parameters do not need to be re-estimated for each domain, making this effective to real use cases. Evaluation is carried out on English-to-French translation for two different testing scenarios. We first consider the case where an end-user performs translations on a known domain. Secondly, we consider the scenario where the domain is not known and predicted at the sentence level before translating. Results show consistent accuracy improvements for both conditions.

arXiv:1709.04685

Machine-Translation History and Evolution: Survey for Arabic-English Translations

As a result of the rapid changes in information and communication technology (ICT), the world has become a small village where people from all over the world connect with each other in dialogue and communication via the Internet. Also, communications have become a daily routine activity due to the new globalization where companies and even universities become global residing cross countries borders. As a result, translation becomes a needed activity in this connected world. ICT made it possible to have a student in one country take a course or even a degree from a different country anytime anywhere easily. The resulted communication still needs a language as a means that helps the receiver understands the contents of the sent message. People need an automated translation application because human translators are hard to find all the times, and the human translations are very expensive comparing to the translations automated process. Several types of research describe the electronic process of the Machine-Translation. In this paper, the authors are going to study some of these previous researches, and they will explore some of the needed tools for the Machine-Translation. This research is going to contribute to the Machine-Translation area by helping future researchers to have a summary for the Machine-Translation groups of research and to let lights on the importance of the translation mechanism.

arXiv:1709.04682

Towards an Arabic-English Machine-Translation Based on Semantic Web

Communication tools make the world like a small village and as a consequence people can contact with others who are from different societies or who speak different languages. This communication cannot happen effectively without Machine Translation because they can be found anytime and everywhere. There are a number of studies that have developed Machine Translation for the English language with so many other languages except the Arabic it has not been considered yet. Therefore we aim to highlight a roadmap for our proposed translation machine to provide an enhanced Arabic English translation based on Semantic.

arXiv:1707.08616

Guiding Reinforcement Learning Exploration Using Natural Language

In this work we present a technique to use natural language to help reinforcement learning generalize to unseen environments. This technique uses neural machine translation, specifically the use of encoder-decoder networks, to learn associations between natural language behavior descriptions and state-action information. We then use this learned model to guide agent exploration using a modified version of policy shaping to make it more effective at learning in unseen environments. We evaluate this technique using the popular arcade game, Frogger, under ideal and non-ideal conditions. This evaluation shows that our modified policy shaping algorithm improves over a Q-learning agent as well as a baseline version of policy shaping.

arXiv:1709.05487

Role of Morphology Injection in Statistical Machine Translation

Phrase-based Statistical models are more commonly used as they perform optimally in terms of both, translation quality and complexity of the system. Hindi and in general all Indian languages are morphologically richer than English. Hence, even though Phrase-based systems perform very well for the less divergent language pairs, for English to Indian language translation, we need more linguistic information (such as morphology, parse tree, parts of speech tags, etc.) on the source side. Factored models seem to be useful in this case, as Factored models consider word as a vector of factors. These factors can contain any information about the surface word and use it while translating. Hence, the objective of this work is to handle morphological inflections in Hindi and Marathi using Factored translation models while translating from English. SMT approaches face the problem of data sparsity while translating into a morphologically rich language. It is very unlikely for a parallel corpus to contain all morphological forms of words. We propose a solution to generate these unseen morphological forms and inject them into original training corpora. In this paper, we study factored models and the problem of sparseness in context of translation to morphologically rich languages. We propose a simple and effective solution which is based on enriching the input with various morphological forms of words. We observe that morphology injection improves the quality of translation in terms of both adequacy and fluency. We verify this with the experiments on two morphologically rich languages: Hindi and Marathi, while translating from English.

arXiv:1701.07274공식 코드 · Papers with Code

Deep Reinforcement Learning: An Overview

We give an overview of recent exciting achievements of deep reinforcement learning (RL). We discuss six core elements, six important mechanisms, and twelve applications. We start with background of machine learning, deep learning and reinforcement learning. Next we discuss core RL elements, including value function, in particular, Deep Q-Network (DQN), policy, reward, model, planning, and exploration. After that, we discuss important mechanisms for RL, including attention and memory, unsupervised learning, transfer learning, multi-agent RL, hierarchical RL, and learning to learn. Then we discuss various applications of RL, including games, in particular, AlphaGo, robotics, natural language processing, including dialogue systems, machine translation, and text generation, computer vision, neural architecture design, business management, finance, healthcare, Industry 4.0, smart grid, intelligent transportation systems, and computer systems. We mention topics not reviewed yet, and list a collection of RL resources. After presenting a brief summary, we close with discussions. Please see Deep Reinforcement Learning, arXiv:1810.06339, for a significant update.

arXiv:1704.00253

Building a Neural Machine Translation System Using Only Synthetic Parallel Data

Recent works have shown that synthetic parallel data automatically generated by translation models can be effective for various neural machine translation (NMT) issues. In this study, we build NMT systems using only synthetic parallel data. As an efficient alternative to real parallel data, we also present a new type of synthetic parallel corpus. The proposed pseudo parallel data are distinct from previous works in that ground truth and synthetic examples are mixed on both sides of sentence pairs. Experiments on Czech-German and French-German translations demonstrate the efficacy of the proposed pseudo parallel corpus, which shows not only enhanced results for bidirectional translation tasks but also substantial improvement with the aid of a ground truth real parallel corpus.

arXiv:1709.05631

Unwritten Languages Demand Attention Too! Word Discovery with Encoder-Decoder Models

Word discovery is the task of extracting words from unsegmented text. In this paper we examine to what extent neural networks can be applied to this task in a realistic unwritten language scenario, where only small corpora and limited annotations are available. We investigate two scenarios: one with no supervision and another with limited supervision with access to the most frequent words. Obtained results show that it is possible to retrieve at least 27% of the gold standard vocabulary by training an encoder-decoder neural machine translation system with only 5,157 sentences. This result is close to those obtained with a task-specific Bayesian nonparametric model. Moreover, our approach has the advantage of generating translation alignments, which could be used to create a bilingual lexicon. As a future perspective, this approach is also well suited to work directly from speech.

arXiv:1709.07809공식 코드 · Papers with Code

Neural Machine Translation

Draft of textbook chapter on neural machine translation. a comprehensive treatment of the topic, ranging from introduction to neural networks, computation graphs, description of the currently dominant attentional sequence-to-sequence model, recent refinements, alternative architectures and challenges. Written as chapter for the textbook Statistical Machine Translation. Used in the JHU Fall 2017 class on machine translation.

arXiv:1709.07758

Improving Language Modelling with Noise-contrastive estimation

Neural language models do not scale well when the vocabulary is large. Noise-contrastive estimation (NCE) is a sampling-based method that allows for fast learning with large vocabularies. Although NCE has shown promising performance in neural machine translation, it was considered to be an unsuccessful approach for language modelling. A sufficient investigation of the hyperparameters in the NCE-based neural language models was also missing. In this paper, we showed that NCE can be a successful approach in neural language modelling when the hyperparameters of a neural network are tuned appropriately. We introduced the 'search-then-converge' learning rate schedule for NCE and designed a heuristic that specifies how to use this schedule. The impact of the other important hyperparameters, such as the dropout rate and the weight initialisation range, was also demonstrated. We showed that appropriate tuning of NCE-based neural language models outperforms the state-of-the-art single-model methods on a popular benchmark.

arXiv:1709.07484

WERd: Using Social Text Spelling Variants for Evaluating Dialectal Speech Recognition

We study the problem of evaluating automatic speech recognition (ASR) systems that target dialectal speech input. A major challenge in this case is that the orthography of dialects is typically not standardized. From an ASR evaluation perspective, this means that there is no clear gold standard for the expected output, and several possible outputs could be considered correct according to different human annotators, which makes standard word error rate (WER) inadequate as an evaluation metric. Such a situation is typical for machine translation (MT), and thus we borrow ideas from an MT evaluation metric, namely TERp, an extension of translation error rate which is closely-related to WER. In particular, in the process of comparing a hypothesis to a reference, we make use of spelling variants for words and phrases, which we mine from Twitter in an unsupervised fashion. Our experiments with evaluating ASR output for Egyptian Arabic, and further manual analysis, show that the resulting WERd (i.e., WER for dialects) metric, a variant of TERp, is more adequate than WER for evaluating dialectal ASR.