arXiv subjects

Haifeng Wang

Publications and source records attributed to Haifeng Wang.

At least 235 records · Page 13Linked to original sources

arXiv2019

Synchronous Speech Recognition and Speech-to-Text Translation with Interactive Decoding

Speech-to-text translation (ST), which translates source language speech into target language text, has attracted intensive attention in recent years. Compared to the traditional pipeline system, the end-to-end ST model has potential benefits of lower latency, smaller model size, and less error propagation. However, it is notoriously difficult to implement such a model without transcriptions as intermediate. Existing works generally apply multi-task learning to improve translation quality by jointly training end-to-end ST along with automatic speech recognition (ASR). However, different tasks in this method cannot utilize information from each other, which limits the improvement. Other works propose a two-stage model where the second model can use the hidden state from the first one, but its cascade manner greatly affects the efficiency of training and inference process. In this paper, we propose a novel interactive attention mechanism which enables ASR and ST to perform synchronously and interactively in a single model. Specifically, the generation of transcriptions and translations not only relies on its previous outputs but also the outputs predicted in the other task. Experiments on TED speech translation corpora have shown that our proposed model can outperform strong baselines on the quality of speech translation and achieve better speech recognition performances as well.

cs.CL↗

arXiv2019

ERNIE 2.0: A Continual Pre-training Framework for Language Understanding

Recently, pre-trained models have achieved state-of-the-art results in various language understanding tasks, which indicates that pre-training on large-scale corpora may play a crucial role in natural language processing. Current pre-training procedures usually focus on training the model with several simple tasks to grasp the co-occurrence of words or sentences. However, besides co-occurring, there exists other valuable lexical, syntactic and semantic information in training corpora, such as named entity, semantic closeness and discourse relations. In order to extract to the fullest extent, the lexical, syntactic and semantic information from training corpora, we propose a continual pre-training framework named ERNIE 2.0 which builds and learns incrementally pre-training tasks through constant multi-task learning. Experimental results demonstrate that ERNIE 2.0 outperforms BERT and XLNet on 16 tasks including English tasks on GLUE benchmarks and several common tasks in Chinese. The source codes and pre-trained models have been released at https://github.com/PaddlePaddle/ERNIE.

cs.CL↗

arXiv2019

Proactive Human-Machine Conversation with Explicit Conversation Goals

Though great progress has been made for human-machine conversation, current dialogue system is still in its infancy: it usually converses passively and utters words more as a matter of response, rather than on its own initiatives. In this paper, we take a radical step towards building a human-like conversational agent: endowing it with the ability of proactively leading the conversation (introducing a new topic or maintaining the current topic). To facilitate the development of such conversation systems, we create a new dataset named DuConv where one acts as a conversation leader and the other acts as the follower. The leader is provided with a knowledge graph and asked to sequentially change the discussion topics, following the given conversation goal, and meanwhile keep the dialogue as natural and engaging as possible. DuConv enables a very challenging task as the model needs to both understand dialogue and plan over the given knowledge graph. We establish baseline results on this dataset (about 270K utterances and 30k dialogues) using several state-of-the-art models. Experimental results show that dialogue models that plan over the knowledge graph can make full use of related knowledge to generate more diverse multi-turn conversations. The baseline systems along with the dataset are publicly available

cs.CL↗

arXiv2019

Model-based Deep Medical Imaging: the roadmap of generalizing iterative reconstruction model using deep learning

Medical imaging is playing a more and more important role in clinics. However, there are several issues in different imaging modalities such as slow imaging speed in MRI, radiation injury in CT and PET. Therefore, accelerating MRI, reducing radiation dose in CT and PET have been ongoing research topics since their invention. Usually, acquiring less data is a direct but important strategy to address these issues. However, less acquisition usually results in aliasing artifacts in reconstructions. Recently, deep learning (DL) has been introduced in medical image reconstruction and shown potential on significantly speeding up MR reconstruction and reducing radiation dose. In this paper, we propose a general framework on combining the reconstruction model with deep learning to maximize the potential of deep learning and model-based reconstruction, and give the examples to demonstrate the performance and requirements of unrolling different algorithms using deep learning.

cs.CV↗

arXiv2019

Knowledge Aware Conversation Generation with Explainable Reasoning over Augmented Graphs

Two types of knowledge, triples from knowledge graphs and texts from documents, have been studied for knowledge aware open-domain conversation generation, in which graph paths can narrow down vertex candidates for knowledge selection decision, and texts can provide rich information for response generation. Fusion of a knowledge graph and texts might yield mutually reinforcing advantages, but there is less study on that. To address this challenge, we propose a knowledge aware chatting machine with three components, an augmented knowledge graph with both triples and texts, knowledge selector, and knowledge aware response generator. For knowledge selection on the graph, we formulate it as a problem of multi-hop graph reasoning to effectively capture conversation flow, which is more explainable and flexible in comparison with previous work. To fully leverage long text information that differentiates our graph from others, we improve a state of the art reasoning algorithm with machine reading comprehension technology. We demonstrate the effectiveness of our system on two datasets in comparison with state-of-the-art models.

cs.AI↗

arXiv2019

Multi-agent Learning for Neural Machine Translation

Conventional Neural Machine Translation (NMT) models benefit from the training with an additional agent, e.g., dual learning, and bidirectional decoding with one agent decoding from left to right and the other decoding in the opposite direction. In this paper, we extend the training framework to the multi-agent scenario by introducing diverse agents in an interactive updating process. At training time, each agent learns advanced knowledge from others, and they work together to improve translation quality. Experimental results on NIST Chinese-English, IWSLT 2014 German-English, WMT 2014 English-German and large-scale Chinese-English translation tasks indicate that our approach achieves absolute improvements over the strong baseline systems and shows competitive performance on all tasks.

cs.CL↗

arXiv2019

DuTongChuan: Context-aware Translation Model for Simultaneous Interpreting

In this paper, we present DuTongChuan, a novel context-aware translation model for simultaneous interpreting. This model allows to constantly read streaming text from the Automatic Speech Recognition (ASR) model and simultaneously determine the boundaries of Information Units (IUs) one after another. The detected IU is then translated into a fluent translation with two simple yet effective decoding strategies: partial decoding and context-aware decoding. In practice, by controlling the granularity of IUs and the size of the context, we can get a good trade-off between latency and translation quality easily. Elaborate evaluation from human translators reveals that our system achieves promising translation quality (85.71% for Chinese-English, and 86.36% for English-Chinese), specially in the sense of surprisingly good discourse coherence. According to an End-to-End (speech-to-speech simultaneous interpreting) evaluation, this model presents impressive performance in reducing latency (to less than 3 seconds at most times). Furthermore, we successfully deploy this model in a variety of Baidu's products which have hundreds of millions of users, and we release it as a service in our AI platform.

cs.CL↗

arXiv2019

Model Learning: Primal Dual Networks for Fast MR imaging

Magnetic resonance imaging (MRI) is known to be a slow imaging modality and undersampling in k-space has been used to increase the imaging speed. However, image reconstruction from undersampled k-space data is an ill-posed inverse problem. Iterative algorithms based on compressed sensing have been used to address the issue. In this work, we unroll the iterations of the primal-dual hybrid gradient algorithm to a learnable deep network architecture, and gradually relax the constraints to reconstruct MR images from highly undersampled k-space data. The proposed method combines the theoretical convergence guarantee of optimi-zation methods with the powerful learning capability of deep networks. As the constraints are gradually relaxed, the reconstruction model is finally learned from the training data by updating in k-space and image domain alternatively. Experi-ments on in vivo MR data demonstrate that the proposed method achieves supe-rior MR reconstructions from highly undersampled k-space data over other state-of-the-art image reconstruction methods.

eess.IV↗

arXiv2019

STACL: Simultaneous Translation with Implicit Anticipation and Controllable Latency using Prefix-to-Prefix Framework

Simultaneous translation, which translates sentences before they are finished, is useful in many scenarios but is notoriously difficult due to word-order differences. While the conventional seq-to-seq framework is only suitable for full-sentence translation, we propose a novel prefix-to-prefix framework for simultaneous translation that implicitly learns to anticipate in a single translation model. Within this framework, we present a very simple yet surprisingly effective wait-k policy trained to generate the target sentence concurrently with the source sentence, but always k words behind. Experiments show our strategy achieves low latency and reasonable quality (compared to full-sentence translation) on 4 directions: zh<->en and de<->en.

cs.CL↗

arXiv2019

Accelerating MR Imaging via Deep Chambolle-Pock Network

Compressed sensing (CS) has been introduced to accelerate data acquisition in MR Imaging. However, CS-MRI methods suffer from detail loss with large acceleration and complicated parameter selection. To address the limitations of existing CS-MRI methods, a model-driven MR reconstruction is proposed that trains a deep network, named CP-net, which is derived from the Chambolle-Pock algorithm to reconstruct the in vivo MR images of human brains from highly undersampled complex k-space data acquired on different types of MR scanners. The proposed deep network can learn the proximal operator and parameters among the Chambolle-Pock algorithm. All of the experiments show that the proposed CP-net achieves more accurate MR reconstruction results, outperforming state-of-the-art methods across various quantitative metrics.

eess.IV↗

arXiv2019

T1rho Fractional-order Relaxation of Human Articular Cartilage

T1rho imaging is a promising non-invasive diagnostic tool for early detection of articular cartilage degeneration. A mono-exponential model is normally used to describe the T1rho relaxation process. However, mono-exponentials may not adequately to describe NMR relaxation in complex, heterogeneous, and anisotropic materials, such as articular cartilage. Fractional-order models have been used successfully to describe complex relaxation phenomena in the laboratory frame in cartilage matrix components. In this paper, we develop a time-fractional order (T-FACT) model for T1rho fitting in human articular cartilage. Representative results demonstrate that the proposed method is able to fit the experimental data with smaller root mean squared error than the one from conventional mono-exponential relaxation model in human articular cartilage.

physics.med-ph↗

arXiv2019

End-to-End Speech Translation with Knowledge Distillation

End-to-end speech translation (ST), which directly translates from source language speech into target language text, has attracted intensive attentions in recent years. Compared to conventional pipeline systems, end-to-end ST models have advantages of lower latency, smaller model size and less error propagation. However, the combination of speech recognition and text translation in one model is more difficult than each of these two tasks. In this paper, we propose a knowledge distillation approach to improve ST model by transferring the knowledge from text translation model. Specifically, we first train a text translation model, regarded as a teacher model, and then ST model is trained to learn output probabilities from teacher model through knowledge distillation. Experiments on English- French Augmented LibriSpeech and English-Chinese TED corpus show that end-to-end ST is possible to implement on both similar and dissimilar language pairs. In addition, with the instruction of teacher model, end-to-end ST model can gain significant improvements by over 3.5 BLEU points.

cs.CL↗

arXiv2019

Application of Time-Fractional Order Bloch Equation in Magnetic Resonance Fingerprinting

Magnetic resonance fingerprinting (MRF) is one novel fast quantitative imaging framework for simultaneous quantification of multiple parameters with pseudo-randomized acquisition patterns. The accuracy of the resulting multi-parameters is very important for clinical applications. In this paper, we derived signal evolutions from the anomalous relaxation using a fractional calculus. More specifically, we utilized time-fractional order extension of the Bloch equations to generate dictionary to provide more complex system descriptions for MRF applications. The representative results of phantom experiments demonstrated the good accuracy performance when applying the time-fractional order Bloch equations to generate dictionary entries in the MRF framework. The utility of the proposed method is also validated by in-vivo study.

physics.med-ph↗

arXiv2019

Fast Calculation Method of Average g-Factor for Wave-CAIPI Imaging

Wave-CAIPI MR imaging is a 3D imaging technique which can uniformize the g-factor maps and significantly reduce g-factor penalty at high acceleration factors. But it is time-consuming to calculate the average g-factor penalty for optimizing the parameters of Wave-CAIPI. In this paper, we propose a novel fast calculation method to calculate the average g-factor in Wave-CAIPI imaging. Wherein, the g-factor value in the arbitrary (e.g. the central) position is separately calculated and then approximated to the average g-factor using Taylor linear approximation. The verification experiments have demonstrated that the average g-factors of Wave-CAIPI imaging which are calculated by the proposed method is consistent with the previous time-consuming theoretical calculation method and the conventional pseudo multiple replica method. Comparison experiments show that the proposed method is averagely about 1000 times faster than the previous theoretical calculation method and about 1700 times faster than the conventional pseudo multiple replica method.

physics.med-ph↗

arXiv2019

A Boost Converter Design with Low Output Ripple Based on Harmonics Feedback

Conventional boost converters are essential to connect low-voltage energy source such as battery with high voltage DC bus in Electric Vehicles due to its simple construction and high conversion efficiency. However, large output capacitor banks must be used in order to reduce the output voltage ripples when an inverter is connected to a boost converter. In a typical series hybrid electric vehicle powertrain, significant voltage and current perturbations will impact on battery system performance and result in considerable power loss. To address this issue, this paper proposes a new control strategy for ripple reduction in the dc-link of power electronic converters. In the proposed method, an observer is designed to adaptively estimate the dc-link voltage and current harmonics. The harmonic terms are multiplied by optimized gains to control the converter's duty cycle by negative feedback law. Entire control model for the proposed converter is illustrated and robustness is verified. Simulation and experiments show that original voltage and current ripple magnitude can be reduced significantly.

eess.SY↗

arXiv2018

Utilizing Imbalanced Data and Classification Cost Matrix to Predict Movie Preferences

In this paper, we propose a movie genre recommendation system based on imbalanced survey data and unequal classification costs for small and medium-sized enterprises (SMEs) who need a data-based and analytical approach to stock favored movies and target marketing to young people. The dataset maintains a detailed personal profile as predictors including demographic, behavioral and preferences information for each user as well as imbalanced genre preferences. These predictors do not include the information such as actors or directors. The paper applies Gentle boost, Adaboost and Bagged tree ensembles as well as SVM machine learning algorithms to learn classification from one thousand observations and predict movie genre preferences with adjusted classification costs. The proposed recommendation system also selects important predictors to avoid overfitting and to shorten training time. This paper compares the test error among the above-mentioned algorithms that are used to recommend different movie genres. The prediction power is also indicated in a comparison of precision and recall with other state-of-the-art recommendation systems. The proposed movie genre recommendation system solves problems such as small dataset, imbalanced response, and unequal classification costs.

cs.IR↗

arXiv2018

Modeling Coherence for Discourse Neural Machine Translation

Discourse coherence plays an important role in the translation of one text. However, the previous reported models most focus on improving performance over individual sentence while ignoring cross-sentence links and dependencies, which affects the coherence of the text. In this paper, we propose to use discourse context and reward to refine the translation quality from the discourse perspective. In particular, we generate the translation of individual sentences at first. Next, we deliberate the preliminary produced translations, and train the model to learn the policy that produces discourse coherent text by a reward teacher. Practical results on multiple discourse test datasets indicate that our model significantly improves the translation quality over the state-of-the-art baseline system by +1.23 BLEU score. Moreover, our model generates more discourse coherent text and obtains +2.2 BLEU improvements when evaluated by discourse metrics.

cs.CL↗

arXiv2018

DuReader: a Chinese Machine Reading Comprehension Dataset from Real-world Applications

This paper introduces DuReader, a new large-scale, open-domain Chinese ma- chine reading comprehension (MRC) dataset, designed to address real-world MRC. DuReader has three advantages over previous MRC datasets: (1) data sources: questions and documents are based on Baidu Search and Baidu Zhidao; answers are manually generated. (2) question types: it provides rich annotations for more question types, especially yes-no and opinion questions, that leaves more opportunity for the research community. (3) scale: it contains 200K questions, 420K answers and 1M documents; it is the largest Chinese MRC dataset so far. Experiments show that human performance is well above current state-of-the-art baseline systems, leaving plenty of room for the community to make improvements. To help the community make these improvements, both DuReader and baseline systems have been posted online. We also organize a shared competition to encourage the exploration of more models. Since the release of the task, there are significant improvements over the baselines.

cs.CL↗