Many machine learning tasks can be expressed as the transformation---or \emph{transduction}---of input sequences into output sequences: speech recognition, machine translation, protein secondary structure prediction and text-to-speech to name but a few. One of the key challenges in sequence transduction is learning to represent both the input and output sequences in a way that is invariant to sequential distortions such as shrinking, stretching and translating. Recurrent neural networks (RNNs) are a powerful sequence learning architecture that has proven capable of learning such representations. However RNNs traditionally require a pre-defined alignment between the input and output sequences to perform transduction. This is a severe limitation since \emph{finding} the alignment is the most difficult aspect of many sequence transduction problems. Indeed, even determining the length of the output sequence is often challenging. This paper introduces an end-to-end, probabilistic sequence transduction system, based entirely on RNNs, that is in principle able to transform any input sequence into any finite, discrete output sequence. Experimental results for phoneme recognition are provided on the TIMIT speech corpus.
Multi-layer models with multiple attention heads per layer provide superior translation quality compared to simpler and shallower models, but determining what source context is most relevant to each target word is more challenging as a result. Therefore, deriving high-accuracy word alignments from the activations of a state-of-the-art neural machine translation model is an open challenge. We propose a simple model extension to the Transformer architecture that makes use of its hidden representations and is restricted to attend solely on encoder information to predict the next word. It can be trained on bilingual data without word-alignment information. We further introduce a novel alignment inference procedure which applies stochastic gradient descent to directly optimize the attention activations towards a given target word. The resulting alignments dramatically outperform the naive approach to interpreting Transformer attention activations, and are comparable to Giza++ on two publicly available data sets.
Recent years has witnessed dramatic progress of neural machine translation (NMT), however, the method of manually guiding the translation procedure remains to be better explored. Previous works proposed to handle such problem through lexcially-constrained beam search in the decoding phase. Unfortunately, these lexically-constrained beam search methods suffer two fatal disadvantages: high computational complexity and hard beam search which generates unexpected translations. In this paper, we propose to learn the ability of lexically-constrained translation with external memory, which can overcome the above mentioned disadvantages. For the training process, automatically extracted phrase pairs are extracted from alignment and sentence parsing, then further be encoded into an external memory. This memory is then used to provide lexically-constrained information for training through a memory-attention machanism. Various experiments are conducted on WMT Chinese to English and English to German tasks. All the results can demonstrate the effectiveness of our method.
The embedding layers transforming input words into real vectors are the key components of deep neural networks used in natural language processing. However, when the vocabulary is large, the corresponding weight matrices can be enormous, which precludes their deployment in a limited resource setting. We introduce a novel way of parametrizing embedding layers based on the Tensor Train (TT) decomposition, which allows compressing the model significantly at the cost of a negligible drop or even a slight gain in performance. We evaluate our method on a wide range of benchmarks in natural language processing and analyze the trade-off between performance and compression ratios for a wide range of architectures, from MLPs to LSTMs and Transformers.
The evaluation of text simplification (TS) systems remains an open challenge. As the task has common points with machine translation (MT), TS is often evaluated using MT metrics such as BLEU. However, such metrics require high quality reference data, which is rarely available for TS. TS has the advantage over MT of being a monolingual task, which allows for direct comparisons to be made between the simplified text and its original version. In this paper, we compare multiple approaches to reference-less quality estimation of sentence-level text simplification systems, based on the dataset used for the QATS 2016 shared task. We distinguish three different dimensions: gram-maticality, meaning preservation and simplicity. We show that n-gram-based MT metrics such as BLEU and METEOR correlate the most with human judgment of grammaticality and meaning preservation, whereas simplicity is best evaluated by basic length-based metrics.
Computations for the softmax function are significantly expensive when the number of output classes is large. In this paper, we present a novel softmax inference speedup method, Doubly Sparse Softmax (DS-Softmax), that leverages sparse mixture of sparse experts to efficiently retrieve top-k classes. Different from most existing methods that require and approximate a fixed softmax, our method is learning-based and can adapt softmax weights for a better inference speedup. In particular, our method learns a two-level hierarchy which divides entire output class space into several partially overlapping experts. Each expert is sparse and only contains a subset of output classes. To find top-k classes, a sparse mixture enables us to find the most probable expert quickly, and the sparse expert enables us to search within a small-scale softmax. We empirically conduct evaluation on several real-world tasks, including neural machine translation, language modeling and image classification, and demonstrate that significant computation reductions can be achieved at no performance loss.
Self-attention is a useful mechanism to build generative models for language and images. It determines the importance of context elements by comparing each element to the current time step. In this paper, we show that a very lightweight convolution can perform competitively to the best reported self-attention results. Next, we introduce dynamic convolutions which are simpler and more efficient than self-attention. We predict separate convolution kernels based solely on the current time-step in order to determine the importance of context elements. The number of operations required by this approach scales linearly in the input length, whereas self-attention is quadratic. Experiments on large-scale machine translation, language modeling and abstractive summarization show that dynamic convolutions improve over strong self-attention models. On the WMT'14 English-German test set dynamic convolutions achieve a new state of the art of 29.7 BLEU.
It is intuitive that NLP tasks for logographic languages like Chinese should benefit from the use of the glyph information in those languages. However, due to the lack of rich pictographic evidence in glyphs and the weak generalization ability of standard computer vision models on character data, an effective way to utilize the glyph information remains to be found. In this paper, we address this gap by presenting Glyce, the glyph-vectors for Chinese character representations. We make three major innovations: (1) We use historical Chinese scripts (e.g., bronzeware script, seal script, traditional Chinese, etc) to enrich the pictographic evidence in characters; (2) We design CNN structures (called tianzege-CNN) tailored to Chinese character image processing; and (3) We use image-classification as an auxiliary task in a multi-task learning setup to increase the model's ability to generalize. We show that glyph-based models are able to consistently outperform word/char ID-based models in a wide range of Chinese NLP tasks. We are able to set new state-of-the-art results for a variety of Chinese NLP tasks, including tagging (NER, CWS, POS), sentence pair classification, single sentence classification tasks, dependency parsing, and semantic role labeling. For example, the proposed model achieves an F1 score of 80.6 on the OntoNotes dataset of NER, +1.5 over BERT; it achieves an almost perfect accuracy of 99.8\% on the Fudan corpus for text classification. Code found at https://github.com/ShannonAI/glyce.
Recent years have seen the rise of Deep Learning (DL) techniques applied to source code. Researchers have exploited DL to automate several development and maintenance tasks, such as writing commit messages, generating comments and detecting vulnerabilities among others. One of the long lasting dreams of applying DL to source code is the possibility to automate non-trivial coding activities. While some steps in this direction have been taken (e.g., learning how to fix bugs), there is still a glaring lack of empirical evidence on the types of code changes that can be learned and automatically applied by DL. Our goal is to make this first important step by quantitatively and qualitatively investigating the ability of a Neural Machine Translation (NMT) model to learn how to automatically apply code changes implemented by developers during pull requests. We train and experiment with the NMT model on a set of 236k pairs of code components before and after the implementation of the changes provided in the pull requests. We show that, when applied in a narrow enough context (i.e., small/medium-sized pairs of methods before/after the pull request changes), NMT can automatically replicate the changes implemented by developers during pull requests in up to 36% of the cases. Moreover, our qualitative analysis shows that the model is capable of learning and replicating a wide variety of meaningful code changes, especially refactorings and bug-fixing activities. Our results pave the way for novel research in the area of DL on code, such as the automatic learning and applications of refactoring.
The parallel corpus for multilingual NLP tasks, deep learning applications like Statistical Machine Translation Systems is very important. The parallel corpus of Hindi-English language pair available for news translation task till date is of very limited size as per the requirement of the systems are concerned. In this work we have developed an automatic parallel corpus generation system prototype, which creates Hindi-English parallel corpus for news translation task. Further to verify the quality of generated parallel corpus we have experimented by taking various performance metrics and the results are quite interesting.
Normalization layers are a staple in state-of-the-art deep neural network architectures. They are widely believed to stabilize training, enable higher learning rate, accelerate convergence and improve generalization, though the reason for their effectiveness is still an active research topic. In this work, we challenge the commonly-held beliefs by showing that none of the perceived benefits is unique to normalization. Specifically, we propose fixed-update initialization (Fixup), an initialization motivated by solving the exploding and vanishing gradient problem at the beginning of training via properly rescaling a standard initialization. We find training residual networks with Fixup to be as stable as training with normalization -- even for networks with 10,000 layers. Furthermore, with proper regularization, Fixup enables residual networks without normalization to achieve state-of-the-art performance in image classification and machine translation.
This review paper discusses how context has been used in neural machine translation (NMT) in the past two years (2017-2018). Starting with a brief retrospect on the rapid evolution of NMT models, the paper then reviews studies that evaluate NMT output from various perspectives, with emphasis on those analyzing limitations of the translation of contextual phenomena. In a subsequent version, the paper will then present the main methods that were proposed to leverage context for improving translation quality, and distinguishes methods that aim to improve the translation of specific phenomena from those that consider a wider unstructured context.
Recently it was shown that linguistic structure predicted by a supervised parser can be beneficial for neural machine translation (NMT). In this work we investigate a more challenging setup: we incorporate sentence structure as a latent variable in a standard NMT encoder-decoder and induce it in such a way as to benefit the translation task. We consider German-English and Japanese-English translation benchmarks and observe that when using RNN encoders the model makes no or very limited use of the structure induction apparatus. In contrast, CNN and word-embedding-based encoders rely on latent graphs and force them to encode useful, potentially long-distance, dependencies.
Recent studies have demonstrated the efficiency of generative pretraining for English natural language understanding. In this work, we extend this approach to multiple languages and show the effectiveness of cross-lingual pretraining. We propose two methods to learn cross-lingual language models (XLMs): one unsupervised that only relies on monolingual data, and one supervised that leverages parallel data with a new cross-lingual language model objective. We obtain state-of-the-art results on cross-lingual classification, unsupervised and supervised machine translation. On XNLI, our approach pushes the state of the art by an absolute gain of 4.9% accuracy. On unsupervised machine translation, we obtain 34.3 BLEU on WMT'16 German-English, improving the previous state of the art by more than 9 BLEU. On supervised machine translation, we obtain a new state of the art of 38.5 BLEU on WMT'16 Romanian-English, outperforming the previous best approach by more than 4 BLEU. Our code and pretrained models will be made publicly available.
Sequence-to-sequence models are commonly trained via maximum likelihood estimation (MLE). However, standard MLE training considers a word-level objective, predicting the next word given the previous ground-truth partial sentence. This procedure focuses on modeling local syntactic patterns, and may fail to capture long-range semantic structure. We present a novel solution to alleviate these issues. Our approach imposes global sequence-level guidance via new supervision based on optimal transport, enabling the overall characterization and preservation of semantic features. We further show that this method can be understood as a Wasserstein gradient flow trying to match our model to the ground truth sequence distribution. Extensive experiments are conducted to validate the utility of the proposed approach, showing consistent improvements over a wide variety of NLP tasks, including machine translation, abstractive text summarization, and image captioning.
Accurate prediction of fading channel in future is essential to realize adaptive transmission and other methods that can save power and provide gains. In practice, wireless channel model can be regarded as a new language model, and the time-varying channel can be seen as mumbling in this language, which is too complex to understand, to say nothing of prediction. Fortunately, neural networks have been proved efficient in learning language models in recent times, moreover, sequence-to-sequence (seq2seq) models can provide the state of the art performance in various tasks such as machine translation, image caption generation, and text summarization. Predicting channel with neural networks seems promising, however, vanilla neural networks cannot deal with complex-valued inputs while channel state information (CSI) is in complex domain. In this paper, we present a powerful method to understand and predict complex-valued channel by utilizing seq2seq models, the results show that seq2seq models are also expert in time series prediction, and realistic channel prediction with comparable or superior performance relative to channel estimation is attainable.
Without real bilingual corpus available, unsupervised Neural Machine Translation (NMT) typically requires pseudo parallel data generated with the back-translation method for the model training. However, due to weak supervision, the pseudo data inevitably contain noises and errors that will be accumulated and reinforced in the subsequent training process, leading to bad translation performance. To address this issue, we introduce phrase based Statistic Machine Translation (SMT) models which are robust to noisy data, as posterior regularizations to guide the training of unsupervised NMT models in the iterative back-translation process. Our method starts from SMT models built with pre-trained language models and word-level translation tables inferred from cross-lingual embeddings. Then SMT and NMT models are optimized jointly and boost each other incrementally in a unified EM framework. In this way, (1) the negative effect caused by errors in the iterative back-translation process can be alleviated timely by SMT filtering noises from its phrase tables; meanwhile, (2) NMT can compensate for the deficiency of fluency inherent in SMT. Experiments conducted on en-fr and en-de translation tasks show that our method outperforms the strong baseline and achieves new state-of-the-art unsupervised machine translation performance.
We present ParaBank, a large-scale English paraphrase dataset that surpasses prior work in both quantity and quality. Following the approach of ParaNMT, we train a Czech-English neural machine translation (NMT) system to generate novel paraphrases of English reference sentences. By adding lexical constraints to the NMT decoding procedure, however, we are able to produce multiple high-quality sentential paraphrases per source sentence, yielding an English paraphrase resource with more than 4 billion generated tokens and exhibiting greater lexical diversity. Using human judgments, we also demonstrate that ParaBank's paraphrases improve over ParaNMT on both semantic similarity and fluency. Finally, we use ParaBank to train a monolingual NMT model with the same support for lexically-constrained decoding for sentence rewriting tasks.
Neural machine translation has significantly pushed forward the quality of the field. However, there are remaining big issues with the output translations and one of them is fairness. Neural models are trained on large text corpora which contain biases and stereotypes. As a consequence, models inherit these social biases. Recent methods have shown results in reducing gender bias in other natural language processing tools such as word embeddings. We take advantage of the fact that word embeddings are used in neural machine translation to propose a method to equalize gender biases in neural machine translation using these representations. Specifically, we propose, experiment and analyze the integration of two debiasing techniques over GloVe embeddings in the Transformer translation architecture. We evaluate our proposed system on the WMT English-Spanish benchmark task, showing gains up to one BLEU point. As for the gender bias evaluation, we generate a test set of occupations and we show that our proposed system learns to equalize existing biases from the baseline system.
Scientific writing is difficult. It is even harder for those for whom English is a second language (ESL learners). Scholars around the world spend a significant amount of time and resources proofreading their work before submitting it for review or publication. In this paper we present a novel machine learning based application for proper word choice task. Proper word choice is a generalization the lexical substitution (LS) and grammatical error correction (GEC) tasks. We demonstrate and evaluate the usefulness of applying bidirectional Long Short Term Memory (LSTM) tagger, for this task. While state-of-the-art grammatical error correction uses error-specific classifiers and machine translation methods, we demonstrate an unsupervised method that is based solely on a high quality text corpus and does not require manually annotated data. We use a bidirectional Recurrent Neural Network (RNN) with LSTM for learning the proper word choice based on a word's sentential context. We demonstrate and evaluate our application on both a domain-specific (scientific), writing task and a general-purpose writing task. We show that our domain-specific and general-purpose models outperform state-of-the-art general context learning. As an additional contribution of this research, we also share our code, pre-trained models, and a new ESL learner test set with the research community.