arXiv subjects

Wei Li

Publications and source records attributed to Wei Li.

At least 1,261 records · Page 70Linked to original sources

arXiv2018

Manipulating orbital angular momentum entanglement by using the Heisenberg Uncertainty principle

Orbital angular momentum entanglement is one of the most intriguing topics in quantum physics. A broad range of research have been dedicated either to unravel its underlying physics or to expand the entanglement dimensions and degrees. In this paper, we present a theoretical study on the orbital angular momentum entanglement by employing the Heisenberg uncertainty principle to quantum position correlation within the azimuthal region. In this study, we decompose the pump light into a set of pump cone states characterized by their radii. The OAM entanglement can be manipulated by controlling the radius of the pump cone state, the length of the nonlinear crystal and the OAM carried by the pump field,which is followed by a detailed discussion. We expect that our research will bring us a deeper understanding of the OAM entanglement, and will do help to the high-dimensional quantum information tasks based on OAM entanglement.

physics.optics↗

arXiv2018

Exploration on Generating Traditional Chinese Medicine Prescription from Symptoms with an End-to-End method

Traditional Chinese Medicine (TCM) is an influential form of medical treatment in China and surrounding areas. In this paper, we propose a TCM prescription generation task that aims to automatically generate a herbal medicine prescription based on textual symptom descriptions. Sequence-to-sequence (seq2seq) model has been successful in dealing with sequence generation tasks. We explore a potential end-to-end solution to the TCM prescription generation task using seq2seq models. However, experiments show that directly applying seq2seq model leads to unfruitful results due to the repetition problem. To solve the problem, we propose a novel decoder with coverage mechanism and a novel soft loss function. The experimental results demonstrate the effectiveness of the proposed approach. Judged by professors who excel in TCM, the generated prescriptions are rated 7.3 out of 10. It shows that the model can indeed help with the prescribing procedure in real life.

cs.CL↗

arXiv2018

Automatic Academic Paper Rating Based on Modularized Hierarchical Convolutional Neural Network

As more and more academic papers are being submitted to conferences and journals, evaluating all these papers by professionals is time-consuming and can cause inequality due to the personal factors of the reviewers. In this paper, in order to assist professionals in evaluating academic papers, we propose a novel task: automatic academic paper rating (AAPR), which automatically determine whether to accept academic papers. We build a new dataset for this task and propose a novel modularized hierarchical convolutional neural network to achieve automatic academic paper rating. Evaluation results show that the proposed model outperforms the baselines by a large margin. The dataset and code are available at \url{https://github.com/lancopku/AAPR}

cs.CL↗

arXiv2018

Adversarial adaptive 1-D convolutional neural networks for bearing fault diagnosis under varying working condition

Traditional intelligent fault diagnosis of rolling bearings work well only under a common assumption that the labeled training data (source domain) and unlabeled testing data (target domain) are drawn from the same distribution. However, in many real-world applications, this assumption does not hold, especially when the working condition varies. In this paper, a new adversarial adaptive 1-D CNN called A2CNN is proposed to address this problem. A2CNN consists of four parts, namely, a source feature extractor, a target feature extractor, a label classifier and a domain discriminator. The layers between the source and target feature extractor are partially untied during the training stage to take both training efficiency and domain adaptation into consideration. Experiments show that A2CNN has strong fault-discriminative and domain-invariant capacity, and therefore can achieve high accuracy under different working conditions. We also visualize the learned features and the networks to explore the reasons behind the high performance of our proposed model.

eess.SP↗

arXiv2018

Appearance-and-Relation Networks for Video Classification

Spatiotemporal feature learning in videos is a fundamental problem in computer vision. This paper presents a new architecture, termed as Appearance-and-Relation Network (ARTNet), to learn video representation in an end-to-end manner. ARTNets are constructed by stacking multiple generic building blocks, called as SMART, whose goal is to simultaneously model appearance and relation from RGB input in a separate and explicit manner. Specifically, SMART blocks decouple the spatiotemporal learning module into an appearance branch for spatial modeling and a relation branch for temporal modeling. The appearance branch is implemented based on the linear combination of pixels or filter responses in each frame, while the relation branch is designed based on the multiplicative interactions between pixels or filter responses across multiple frames. We perform experiments on three action recognition benchmarks: Kinetics, UCF101, and HMDB51, demonstrating that SMART blocks obtain an evident improvement over 3D convolutions for spatiotemporal feature learning. Under the same training setting, ARTNets achieve superior performance on these three datasets to the existing state-of-the-art methods.

cs.CV↗

arXiv2018

Monolithic quantum-dot distributed feedback laser array on silicon

Electrically-pumped lasers directly grown on silicon are key devices interfacing silicon microelectronics and photonics. We report here, for the first time, an electrically-pumped, room-temperature, continuous-wave (CW) and single-mode distributed feedback (DFB) laser array fabricated in InAs/GaAs quantum-dot (QD) gain material epitaxially grown on silicon. CW threshold currents as low as 12 mA and single-mode side mode suppression ratios (SMSRs) as high as 50 dB have been achieved from individual devices in the array. The laser array, compatible with state-of-the-art coarse wavelength division multiplexing (CWDM) systems, has a well-aligned channel spacing of 20 0.2 nm and exhibits a record wavelength coverage range of 100 nm, the full span of the O-band. These results indicate that, for the first time, the performance of lasers epitaxially grown on silicon is elevated to a point approaching real-world CWDM applications, demonstrating the great potential of this technology.

physics.optics↗

arXiv2018

Generative Model for Heterogeneous Inference

Generative models (GMs) such as Generative Adversary Network (GAN) and Variational Auto-Encoder (VAE) have thrived these years and achieved high quality results in generating new samples. Especially in Computer Vision, GMs have been used in image inpainting, denoising and completion, which can be treated as the inference from observed pixels to corrupted pixels. However, images are hierarchically structured which are quite different from many real-world inference scenarios with non-hierarchical features. These inference scenarios contain heterogeneous stochastic variables and irregular mutual dependences. Traditionally they are modeled by Bayesian Network (BN). However, the learning and inference of BN model are NP-hard thus the number of stochastic variables in BN is highly constrained. In this paper, we adapt typical GMs to enable heterogeneous learning and inference in polynomial time.We also propose an extended autoregressive (EAR) model and an EAR with adversary loss (EARA) model and give theoretical results on their effectiveness. Experiments on several BN datasets show that our proposed EAR model achieves the best performance in most cases compared to other GMs. Except for black box analysis, we've also done a serial of experiments on Markov border inference of GMs for white box analysis and give theoretical results.

cs.LG↗

arXiv2018

Anisotropic superconductivity and elongated vortices with unusual bound states in quasi-one-dimensional nickel-bismuth compounds

We report low temperature scanning tunneling microscopy and spectroscopy studies of Ni-Bi films grown by molecular beam epitaxy. Highly anisotropic and twofold symmetric superconducting gaps are revealed in two distinct composites, Bi-rich NiBi3 and near-equimolar NixBi, both sharing quasi-one-dimensional crystal structure. We further reveal axially elongated vortices in both phases, but Caroli-de Gennes-Matricon states solely within the vortex cores of NiBi3. Intriguingly, although the localized bound state splits energetically off at a finite distance ~10 nm away from a vortex center along the minor axis of elliptic vortex, no splitting is found along the major axis. We attribute the elongated vortices and unusual vortex behaviors to the combined effects of twofold superconducting gap and Fermi velocity. The findings provide a comprehensive understanding of the electron pairing and vortex matter in quasi-one-dimensional superconductors

cond-mat.supr-con↗

arXiv2018

Monocular Vision-based Vehicle Localization Aided by Fine-grained Classification

Monocular camera systems are prevailing in intelligent transportation systems, but by far they have rarely been used for dimensional purposes such as to accurately estimate the localization information of a vehicle. In this paper, we show that this capability can be realized. By integrating a series of advanced computer vision techniques including foreground extraction, edge and line detection, etc., and by utilizing deep learning networks for fine-grained vehicle model classification, we developed an algorithm which can estimate vehicles location (position, orientation and boundaries) within the environment down to 3.79 percent position accuracy and 2.5 degrees orientation accuracy. With this enhancement, current massive surveillance camera systems can potentially play the role of e-traffic police and trigger many new intelligent transportation applications, for example, to guide vehicles for parking or even for autonomous driving.

cs.RO↗

arXiv2018

A hybrid inverse problem in the fluorescence ultrasound modulated optical tomography in the diffusive regime

We investigate a hybrid inverse problem in fluorescence ultrasound modulated optical tomography (fUMOT) in the diffusive regime. We prove that the absorption coefficient of the fluorophores at the excitation frequency and the quantum efficiency coefficient can be uniquely and stably reconstructed from boundary measurement of the photon currents, provided that some background medium parameters are known. Reconstruction algorithms are proposed and numerically implemented as well.

math.AP↗

arXiv2018

Query and Output: Generating Words by Querying Distributed Word Representations for Paraphrase Generation

Most recent approaches use the sequence-to-sequence model for paraphrase generation. The existing sequence-to-sequence model tends to memorize the words and the patterns in the training dataset instead of learning the meaning of the words. Therefore, the generated sentences are often grammatically correct but semantically improper. In this work, we introduce a novel model based on the encoder-decoder framework, called Word Embedding Attention Network (WEAN). Our proposed model generates the words by querying distributed word representations (i.e. neural word embeddings), hoping to capturing the meaning of the according words. Following previous work, we evaluate our model on two paraphrase-oriented tasks, namely text simplification and short text abstractive summarization. Experimental results show that our model outperforms the sequence-to-sequence baseline by the BLEU score of 6.3 and 5.5 on two English text simplification datasets, and the ROUGE-2 F1 score of 5.7 on a Chinese summarization dataset. Moreover, our model achieves state-of-the-art performances on these three benchmark datasets.

cs.CL↗

arXiv2018

Transferable Joint Attribute-Identity Deep Learning for Unsupervised Person Re-Identification

Most existing person re-identification (re-id) methods require supervised model learning from a separate large set of pairwise labelled training data for every single camera pair. This significantly limits their scalability and usability in real-world large scale deployments with the need for performing re-id across many camera views. To address this scalability problem, we develop a novel deep learning method for transferring the labelled information of an existing dataset to a new unseen (unlabelled) target domain for person re-id without any supervised learning in the target domain. Specifically, we introduce an Transferable Joint Attribute-Identity Deep Learning (TJ-AIDL) for simultaneously learning an attribute-semantic and identitydiscriminative feature representation space transferrable to any new (unseen) target domain for re-id tasks without the need for collecting new labelled training data from the target domain (i.e. unsupervised learning in the target domain). Extensive comparative evaluations validate the superiority of this new TJ-AIDL model for unsupervised person re-id over a wide range of state-of-the-art methods on four challenging benchmarks including VIPeR, PRID, Market-1501, and DukeMTMC-ReID.

cs.CV↗

arXiv2018

SeqFace: Make full use of sequence information for face recognition

Deep convolutional neural networks (CNNs) have greatly improved the Face Recognition (FR) performance in recent years. Almost all CNNs in FR are trained on the carefully labeled datasets containing plenty of identities. However, such high-quality datasets are very expensive to collect, which restricts many researchers to achieve state-of-the-art performance. In this paper, we propose a framework, called SeqFace, for learning discriminative face features. Besides a traditional identity training dataset, the designed SeqFace can train CNNs by using an additional dataset which includes a large number of face sequences collected from videos. Moreover, the label smoothing regularization (LSR) and a new proposed discriminative sequence agent (DSA) loss are employed to enhance discrimination power of deep face features via making full use of the sequence data. Our method achieves excellent performance on Labeled Faces in the Wild (LFW), YouTube Faces (YTF), only with a single ResNet. The code and models are publicly available on-line (https://github.com/huangyangyu/SeqFace).

cs.CV↗

arXiv2018

Robotic Sewing and Knot Tying for Personalized Stent Graft Manufacturing

This paper presents a versatile robotic system for sewing 3D structured object. Leveraging on using a customized robotic sewing device and closed-loop visual servoing control, an all-in-one solution for sewing personalized stent graft is demonstrated. Stitch size planning and automatic knot tying are proposed as the two key functions of the system. By using effective stitch size planning, sub-millimetre sewing accuracy is achieved for stitch sizes ranging from 2mm to 5mm. In addition, a thread manipulator for thread management and tension control is also proposed to perform successive knot tying to secure each stitch. Detailed laboratory experiments have been performed to access the proposed instruments and allied algorithms. The proposed framework can be generalised to a wide range of applications including 3D industrial sewing, as well as transferred to other clinical areas such as surgical suturing.

cs.RO↗

arXiv2018

A switchable diode based on room-temperature two-dimensional ferroelectric α-In2Se3 thin layers

Nanoscaled room-temperature ferroelectricity is ideal for developing advanced non-volatile high-density memories. However, reaching the thin film limit in conventional ferroelectrics is a long-standing challenge due to the possible critical thickness effect. Van der Waals materials, thanks to their stable layered structure, saturate interfacial chemistry and weak interlayer couplings, are promising for exploring ultra-thin two-dimensional (2D) ferroelectrics and device applications. Here, we demonstrate a switchable room-temperature ferroelectric diode built upon a 2D ferroelectric α-In2Se3 layer as thin as 5 nm in the form of graphene/α-In2Se3 heterojunction. The intrinsic out-of-plane ferroelectricity of the α-In2Se3 thin layers is evidenced by the observation of reversible spontaneous electric polarization with a relative low coercive electric field of ~$2 X 10^5 V/cm$ and a typical ferroelectric domain size of around tens $μm^2$. Owing to the out-of-plane ferroelectricity of the α-In2Se3 layer, the Schottky barrier at the graphene/α-In2Se3 interface can be effectively tuned by switching the electric polarization with an applied voltage, leading to a pronounced switchable double diode effect with an on/off ratio of ~$10^4$. Our results offer a new way for developing novel nanoelectronic devices based on 2D ferroelectrics.

cond-mat.mtrl-sci↗

arXiv2018

Emergent spin-1 trimerized valence bond crystal in the spin-1/2 Heisenberg model on the star lattice

We explore the frustrated spin-$1/2$ Heisenberg model on the star lattice with antiferromagnetic (AF) couplings inside each triangle and ferromagnetic (FM) inter-triangle couplings ($J_e<0$), and calculate its magnetic and thermodynamic properties. We show that the FM couplings do not sabotage the magnetic disordering of the ground state due to the frustration from the AF interactions inside each triangle, but trigger a fully gapped inversion-symmetry-breaking trimerized valence bond crystal (TVBC) with emergent spin-1 degrees of freedom. We discover that with strengthening $J_e$, the system scales exponentially, either with or without a magnetic field $h$: the order parameter, the five critical fields that separate the $J_e$-$h$ ground-state phase diagram into six phases, and the excitation gap obtained by low-temperature specific heat, all depend exponentially on $J_e$. We calculate the temperature dependence of the specific heat, which can be directly compared with future experiments.

cond-mat.str-el↗

arXiv2018

Observation of Interface Superconductivity in a SnSe2-Epitaxial Graphene van der Waals Heterostructure

We report on the direct observation of interface superconductivity in single-unit-cell SnSe2 films grown on graphitized SiC(0001) substrate by means of van der Waals epitaxy. Tunneling spectrum in the superconducting state reveals rather conventional character with a fully gapped order parameter. The occurrence of superconductivity is further confirmed by the presence of vortices under external magnetic field. Through interface engineering, we unravel the mechanism of superconductivity that originates from a two-dimensional electron gas formed at the interface of SnSe2 and graphene. Our finding opens up novel strategies to hunt for and understand interface superconductivity based on van der Waals heterostructures.

cond-mat.supr-con↗

arXiv2018

Harmonious Attention Network for Person Re-Identification

Existing person re-identification (re-id) methods either assume the availability of well-aligned person bounding box images as model input or rely on constrained attention selection mechanisms to calibrate misaligned images. They are therefore sub-optimal for re-id matching in arbitrarily aligned person images potentially with large human pose variations and unconstrained auto-detection errors. In this work, we show the advantages of jointly learning attention selection and feature representation in a Convolutional Neural Network (CNN) by maximising the complementary information of different levels of visual attention subject to re-id discriminative learning constraints. Specifically, we formulate a novel Harmonious Attention CNN (HA-CNN) model for joint learning of soft pixel attention and hard regional attention along with simultaneous optimisation of feature representations, dedicated to optimise person re-id in uncontrolled (misaligned) images. Extensive comparative evaluations validate the superiority of this new HA-CNN model for person re-id over a wide variety of state-of-the-art methods on three large-scale benchmarks including CUHK03, Market-1501, and DukeMTMC-ReID.

cs.CV↗