arXiv subjects

Chengyu Wang

Publications and source records attributed to Chengyu Wang.

125 records · Page 7Linked to original sources

arXiv2021

Snapshot Ptychography on Array cameras

We use convolutional neural networks to recover images optically down-sampled by $6.7\times$ using coherent aperture synthesis over a 16 camera array. Where conventional ptychography relies on scanning and oversampling, here we apply decompressive neural estimation to recover full resolution image from a single snapshot, although as shown in simulation multiple snapshots can be used to improve SNR. In place training on experimental measurements eliminates the need to directly calibrate the measurement system. We also present simulations of diverse array camera sampling strategies to explore how snapshot compressive systems might be optimized.

eess.IV↗

arXiv2021

Path-Enhanced Multi-Relational Question Answering with Knowledge Graph Embeddings

The multi-relational Knowledge Base Question Answering (KBQA) system performs multi-hop reasoning over the knowledge graph (KG) to achieve the answer. Recent approaches attempt to introduce the knowledge graph embedding (KGE) technique to handle the KG incompleteness but only consider the triple facts and neglect the significant semantic correlation between paths and multi-relational questions. In this paper, we propose a Path and Knowledge Embedding-Enhanced multi-relational Question Answering model (PKEEQA), which leverages multi-hop paths between entities in the KG to evaluate the ambipolar correlation between a path embedding and a multi-relational question embedding via a customizable path representation mechanism, benefiting for achieving more accurate answers from the perspective of both the triple facts and the extra paths. Experimental results illustrate that PKEEQA improves KBQA models' performance for multi-relational question answering with explainability to some extent derived from paths.

cs.CL↗

arXiv2021

Knowledge-Empowered Representation Learning for Chinese Medical Reading Comprehension: Task, Model and Resources

Machine Reading Comprehension (MRC) aims to extract answers to questions given a passage. It has been widely studied recently, especially in open domains. However, few efforts have been made on closed-domain MRC, mainly due to the lack of large-scale training data. In this paper, we introduce a multi-target MRC task for the medical domain, whose goal is to predict answers to medical questions and the corresponding support sentences from medical information sources simultaneously, in order to ensure the high reliability of medical knowledge serving. A high-quality dataset is manually constructed for the purpose, named Multi-task Chinese Medical MRC dataset (CMedMRC), with detailed analysis conducted. We further propose the Chinese medical BERT model for the task (CMedBERT), which fuses medical knowledge into pre-trained language models by the dynamic fusion mechanism of heterogeneous features and the multi-task learning strategy. Experiments show that CMedBERT consistently outperforms strong baselines by fusing context-aware and knowledge-aware token representations.

cs.CL↗

arXiv2021

EasyTransfer -- A Simple and Scalable Deep Transfer Learning Platform for NLP Applications

The literature has witnessed the success of leveraging Pre-trained Language Models (PLMs) and Transfer Learning (TL) algorithms to a wide range of Natural Language Processing (NLP) applications, yet it is not easy to build an easy-to-use and scalable TL toolkit for this purpose. To bridge this gap, the EasyTransfer platform is designed to develop deep TL algorithms for NLP applications. EasyTransfer is backended with a high-performance and scalable engine for efficient training and inference, and also integrates comprehensive deep TL algorithms, to make the development of industrial-scale TL applications easier. In EasyTransfer, the built-in data and model parallelism strategies, combined with AI compiler optimization, show to be 4.0x faster than the community version of distributed training. EasyTransfer supports various NLP models in the ModelZoo, including mainstream PLMs and multi-modality models. It also features various in-house developed TL algorithms, together with the AppZoo for NLP applications. The toolkit is convenient for users to quickly start model training, evaluation, and online deployment. EasyTransfer is currently deployed at Alibaba to support a variety of business scenarios, including item recommendation, personalized search, conversational question answering, etc. Extensive experiments on real-world datasets and online applications show that EasyTransfer is suitable for online production with cutting-edge performance for various applications. The source code of EasyTransfer is released at Github (https://github.com/alibaba/EasyTransfer).

cs.CL↗

arXiv2021

SMedBERT: A Knowledge-Enhanced Pre-trained Language Model with Structured Semantics for Medical Text Mining

Recently, the performance of Pre-trained Language Models (PLMs) has been significantly improved by injecting knowledge facts to enhance their abilities of language understanding. For medical domains, the background knowledge sources are especially useful, due to the massive medical terms and their complicated relations are difficult to understand in text. In this work, we introduce SMedBERT, a medical PLM trained on large-scale medical corpora, incorporating deep structured semantic knowledge from neighbors of linked-entity.In SMedBERT, the mention-neighbor hybrid attention is proposed to learn heterogeneous-entity information, which infuses the semantic representations of entity types into the homogeneous neighboring entity structure. Apart from knowledge integration as external features, we propose to employ the neighbors of linked-entities in the knowledge graph as additional global contexts of text mentions, allowing them to communicate via shared neighbors, thus enrich their semantic representations. Experiments demonstrate that SMedBERT significantly outperforms strong baselines in various knowledge-intensive Chinese medical tasks. It also improves the performance of other tasks such as question answering, question matching and natural language inference.

cs.CL↗

arXiv2020

Multiscale Phase Retrieval

While characterization of coherent wavefields is essential to laser, x-ray and electron imaging, sensors measure the squared magnitude of the field, rather than the field itself. Holography or phase retrieval must be used to characterize the field. The need for a reference severely restricts the utility of holography. Phase retrieval, in contrast, is theoretically consistent with sensors that directly measure coherent or partially coherent fields with no prior assumptions. Unfortunately, phase retrieval has not yet been successfully implemented for large-scale fields. Here we show that both holography and phase retrieval are capable of quantum-limited coherent signal estimation and we describe phase retrieval strategies that approach the quantum limit for >1 megapixel fields. These strategies rely on group testing using networks of interferometers, such as might be constructed using emerging integrated photonic, plasmonic and/or metamaterial devices. Phase-sensitive sensor planes using such devices could eliminate the need both for lenses and reference signals, creating a path to large aperture diffraction limited laser imaging.

eess.IV↗

arXiv2020

KEML: A Knowledge-Enriched Meta-Learning Framework for Lexical Relation Classification

Lexical relations describe how concepts are semantically related, in the form of relation triples. The accurate prediction of lexical relations between concepts is challenging, due to the sparsity of patterns indicating the existence of such relations. We propose the Knowledge-Enriched Meta-Learning (KEML) framework to address the task of lexical relation classification. In KEML, the LKB-BERT (Lexical Knowledge Base-BERT) model is presented to learn concept representations from massive text corpora, with rich lexical knowledge injected by distant supervision. A probabilistic distribution of auxiliary tasks is defined to increase the model's ability to recognize different types of lexical relations. We further combine a meta-learning process over the auxiliary task distribution and supervised learning to train the neural lexical relation classifier. Experiments over multiple datasets show that KEML outperforms state-of-the-art methods.

cs.CL↗

arXiv2020

INT8 Winograd Acceleration for Conv1D Equipped ASR Models Deployed on Mobile Devices

The intensive computation of Automatic Speech Recognition (ASR) models obstructs them from being deployed on mobile devices. In this paper, we present a novel quantized Winograd optimization pipeline, which combines the quantization and fast convolution to achieve efficient inference acceleration on mobile devices for ASR models. To avoid the information loss due to the combination of quantization and Winograd convolution, a Range-Scaled Quantization (RSQ) training method is proposed to expand the quantized numerical range and to distill knowledge from high-precision values. Moreover, an improved Conv1D equipped DFSMN (ConvDFSMN) model is designed for mobile deployment. We conduct extensive experiments on both ConvDFSMN and Wav2letter models. Results demonstrate the models can be effectively optimized with the proposed pipeline. Especially, Wav2letter achieves 1.48* speedup with an approximate 0.07% WER decrease on ARMv7-based mobile devices.

cs.SD↗

arXiv2020

EasyASR: A Distributed Machine Learning Platform for End-to-end Automatic Speech Recognition

We present EasyASR, a distributed machine learning platform for training and serving large-scale Automatic Speech Recognition (ASR) models, as well as collecting and processing audio data at scale. Our platform is built upon the Machine Learning Platform for AI of Alibaba Cloud. Its main functionality is to support efficient learning and inference for end-to-end ASR models on distributed GPU clusters. It allows users to learn ASR models with either pre-defined or user-customized network architectures via simple user interface. On EasyASR, we have produced state-of-the-art results over several public datasets for Mandarin speech recognition.

cs.CL↗

arXiv2020

Weakly Supervised Construction of ASR Systems with Massive Video Data

Building Automatic Speech Recognition (ASR) systems from scratch is significantly challenging, mostly due to the time-consuming and financially-expensive process of annotating a large amount of audio data with transcripts. Although several unsupervised pre-training models have been proposed, applying such models directly might still be sub-optimal if more labeled, training data could be obtained without a large cost. In this paper, we present a weakly supervised framework for constructing ASR systems with massive video data. As videos often contain human-speech audios aligned with subtitles, we consider videos as an important knowledge source, and propose an effective approach to extract high-quality audios aligned with transcripts from videos based on Optical Character Recognition (OCR). The underlying ASR model can be fine-tuned to fit any domain-specific target training datasets after weakly supervised pre-training. Extensive experiments show that our framework can easily produce state-of-the-art results on six public datasets for Mandarin speech recognition.

eess.AS↗

arXiv2020

Meta Fine-Tuning Neural Language Models for Multi-Domain Text Mining

Pre-trained neural language models bring significant improvement for various NLP tasks, by fine-tuning the models on task-specific training sets. During fine-tuning, the parameters are initialized from pre-trained models directly, which ignores how the learning process of similar NLP tasks in different domains is correlated and mutually reinforced. In this paper, we propose an effective learning procedure named Meta Fine-Tuning (MFT), served as a meta-learner to solve a group of similar NLP tasks for neural language models. Instead of simply multi-task training over all the datasets, MFT only learns from typical instances of various domains to acquire highly transferable knowledge. It further encourages the language model to encode domain-invariant representations by optimizing a series of novel domain corruption loss functions. After MFT, the model can be fine-tuned for each domain with better parameter initializations and higher generalization ability. We implement MFT upon BERT to solve several multi-domain text mining tasks. Experimental results confirm the effectiveness of MFT and its usefulness for few-shot learning.

cs.CL↗

arXiv2020

Intelligent Autofocus

We demonstrate that deep learning methods can determine the best focus position from 1-2 image samples, enabling 5-10x faster focus than traditional search-based methods. In contrast with phase detection methods, deep autofocus does not require specialized hardware. In further constrast with conventional methods, which assume a static "best focus," AI methods can generate scene-based focus trajectories that optimize synthesized image quality for dynamic and three dimensional scenes.

eess.IV↗

arXiv2020

Smart Cameras

We review camera architecture in the age of artificial intelligence. Modern cameras use physical components and software to capture, compress and display image data. Over the past 5 years, deep learning solutions have become superior to traditional algorithms for each of these functions. Deep learning enables 10-100x reduction in electrical sensor power per pixel, 10x improvement in depth of field and dynamic range and 10-100x improvement in image pixel count. Deep learning enables multiframe and multiaperture solutions that fundamentally shift the goals of physical camera design. Here we review the state of the art of deep learning in camera operations and consider the impact of AI on the physical design of cameras.

eess.IV↗

arXiv2020

Population pharmacokinetics and dosing regimen optimization of tacrolimus in Chinese lung transplant recipients

We aimed to develop a population pharmacokinetic model of tacrolimus in Chinese lung transplant recipients, and propose model based dosing regimens for individualized treatment. We obtained 807 tacrolimus whole blood concentrations from 52 lung transplant patients and genotyped CYP3A5*3. Population pharmacokinetic analysis was performed using nonlinear mixed effects modeling. Monte Carlo simulations were employed to design initial dosing regimens. Tacrolimus pharmacokinetics was described by a one compartment model with first order absorption and elimination process. The mean estimated apparent clearance was 13.1 l/h with 20.1% inter subject variability in CYP3A5*3/*3 70kg patients with 30% hematocrit and voriconazole free therapy, which is lower than that in Caucasian(17.5 to 36.5 l/h). Hematocrit, postoperative days, tacrolimus daily dose, voriconazole cotherapy, and CYP3A5*3 genotype were identified as significant covariates for tacrolimus clearance. To achieve the target trough concentration (10 to 15 ng/ml) on the 8th day after transplantation, CYP3A5*1/*3 patients with voriconazole free cotherapy, a higher initial dosage than the current regimen of 0.04 mg/kg q12h should be recommened. Given the nonlinear kinetics of tacrolimus and large variability, population pharmacokinetic model should be combined with therapeutic drug monitoring to optimize individualized therapy.

q-bio.TO↗

arXiv2019

Video Generation from Single Semantic Label Map

This paper proposes the novel task of video generation conditioned on a SINGLE semantic label map, which provides a good balance between flexibility and quality in the generation process. Different from typical end-to-end approaches, which model both scene content and dynamics in a single step, we propose to decompose this difficult task into two sub-problems. As current image generation methods do better than video generation in terms of detail, we synthesize high quality content by only generating the first frame. Then we animate the scene based on its semantic meaning to obtain the temporally coherent video, giving us excellent results overall. We employ a cVAE for predicting optical flow as a beneficial intermediate step to generate a video sequence conditioned on the initial single frame. A semantic label map is integrated into the flow prediction module to achieve major improvements in the image-to-video generation process. Extensive experiments on the Cityscapes dataset show that our method outperforms all competing methods.

cs.CV↗

arXiv2019

S-type Negative Differential Resistance in Semiconducting Transition-Metal Dichalcogenides

Current-controlled (also known as "S-type") negative differential resistance (NDR) is of crucial importance to many emerging applications including neuromorphic computing and high-density memristors integration. However, the experimental realization of S-type NDR based on conventional mechanisms poses demanding requirements on materials, which greatly limits their potential applications. Here, we experimentally identify that semiconducting transition metal dichalcogenides (TMDs) can host a bipolar S-type NDR devices. Theoretical simulations indicate that the origin of the NDR in these devices arises from a thermal feedback mechanism. Furthermore, we demonstrate the potential applications of TMDs based S-type NDR device in signal processing and neuromorphic electronics.

cond-mat.mes-hall↗

arXiv2018

Experimental identification of critical condition for drastically enhancing thermoelectric power factor of two-dimensional layered materials

Nano-structuring is an extremely promising path to high performance thermoelectrics. Favorable improvements in thermal conductivity are attainable in many material systems, and theoretical work points to large improvements in electronic properties. However, realization of the electronic benefits in practical materials has been elusive experimentally. A key challenge is that experimental identification of the quantum confinement length, below which the thermoelectric power factor is significantly enhanced, remains elusive due to lack of simultaneous control of size and carrier density. Here we investigate gate tunable and temperature-dependent thermoelectric transport in $γ$ phase indium selenide ($γ$ InSe, n type semiconductor) samples with thickness varying from 7 to 29 nm. This allows us to properly map out dimension and doping space. Combining theoretical and experimental studies, we reveal that the sharper pre-edge of the conduction-band density of states arising from quantum confinement gives rise to an enhancement of the Seebeck coefficient and the power factor in the thinner InSe samples. Most importantly, we experimentally identify the role of the competition between quantum confinement length and thermal de Broglie wavelength in the enhancement of power factor. Our results provide an important and general experimental guideline for optimizing the power factor and improving the thermoelectric performance of two-dimensional layered semiconductors.

cond-mat.mes-hall↗