arXiv subjects

Wei Xu

Publications and source records attributed to Wei Xu.

At least 649 records · Page 36Linked to original sources

arXiv2016

Role of valence changes and nanoscale atomic displacements in BiS2-based superconductors

Superconductivity within layered crystal structures has attracted sustained interest among condensed matter community, primarily due to their exotic superconducting properties. EuBiS2F is a newly discovered member in the BiS2-based superconducting family, which shows superconductivity at 0.3 K without extrinsic doping. With 50 at.% Ce substitution for Eu, superconductivity is enhanced with Tc increased up to 2.2 K. However, the mechanisms for the Tc enhancement have not yet been elucidated. In this study, the Ce-doping effect on the self-electron-doped superconductor EuBiS2F was investigated by X-ray absorption spectroscopy (XAS). We have established a relationship between Ce-doping and the Tc enhancement in terms of Eu valence changes and nanoscale atomic displacements. The new finding sheds light on the interplay among superconductivity, charge and local structure in BiS2-based superconductors.

cond-mat.str-el↗

arXiv2016

La-doping effect on spin-orbit coupled Sr2IrO4 probed by x-ray absorption spectroscopy

Sr2IrO4 was predicted to be an unconventional superconductor upon carrier doping since it highly resembles the high-temperature cuprates. Here, to understand carrier doping effect on spin-orbit coupled Mott insulator Sr2IrO4, the electronic structure and local structure distortion for Sr2-xLaxIrO4 system have been investigated by x-ray absorption spectroscopy (XAS). By comparing the intensity of white-line features at the Ir L2,3 absorption edges, we observe remarkably large branching ratios in La-doped compounds, greater than that of the parent material Sr2IrO4, suggesting a strong spin-orbit interaction (SOI) for Sr2IrO4-based system. Moreover, extended x-ray absorption fine structure (EXAFS) spectra demonstrate more regular IrO6 octahedra, i.e. the weakened crystal electric field (CEF) versus La-doping. By theoretical calculations, the synergistic effect of regular IrO6 octahedra and electron doping is established, which accounts for the transition from a Mott insulator to a conductive state in Sr2-xLaxIrO4-based system.

cond-mat.str-el↗

arXiv2016

Secure Massive MIMO Systems with Limited RF Chains

In future practical deployments of massive multi-input multi-output (MIMO) systems, the number of radio frequency (RF) chains at the base stations (BSs) may be much smaller than the number of BS antennas to reduce the overall expenditure. In this paper, we propose a novel design framework for joint data and artificial noise (AN) precoding in a multiuser massive MIMO system with limited number of RF chains, which improves the wireless security performance. With imperfect channel state information (CSI), we analytically derive an achievable lower bound on the ergodic secrecy rate of any mobile terminal (MT), for both analog and hybrid precoding schemes. The closed-form lower bound is used to determine optimal power splitting between data and AN that maximizes the secrecy rate through simple one-dimensional search. Analytical and numerical results together reveal that the proposed hybrid precoder, although suffers from reduced secrecy rate compared with theoretical full-dimensional precoder, is free of the high computational complexity of large-scale matrix inversion and null-space calculations, and largely reduces the hardware cost.

cs.IT↗

arXiv2016

Multiband Effects and the Bose-Hubbard Model in One-Dimensional Lattices

We study phase diagrams of one-dimensional bosons with contact interactions in the presence of a lattice. We use the worm algorithm in continuous space and focus on the incommensurate superfluid Mott-insulator transition. Our results are compared to those from the one-band Bose-Hubbard model. When Wannier states are used to determine the Bose-Hubbard model parameters, the comparison unveils an apparent breakdown of the one-band description for strong interactions, even for the Mott-insulating state with an average of one particle per site ($n=1$) in deep lattices. We introduce an inverse confined scattering analysis to obtain the ratio $U/J$, with which the Bose-Hubbard model provides correct results for strong interactions, deep lattices, and $n=1$.

cond-mat.quant-gas↗

arXiv2016

Universal scaling of density and momentum distributions in Lieb-Liniger gases

We present an exact numerical study of the scaling of density and momentum distribution functions of harmonically trapped one-dimensional bosons with repulsive contact interactions at zero and finite temperatures. We use path integral quantum Monte Carlo with worm updates in our calculations at finite interaction strengths, and the Bose-Fermi mapping in the Tonks-Girardeau regime. We discuss the homogeneous case and, within the local density approximation, use it to motivate the scaling in the presence of a harmonic trap. For the momentum distribution function, we pay special attention to the high momentum tails and their $k^{-4}$ asymptotic behavior.

cond-mat.quant-gas↗

arXiv2016

Dataset and Neural Recurrent Sequence Labeling Model for Open-Domain Factoid Question Answering

While question answering (QA) with neural network, i.e. neural QA, has achieved promising results in recent years, lacking of large scale real-word QA dataset is still a challenge for developing and evaluating neural QA system. To alleviate this problem, we propose a large scale human annotated real-world QA dataset WebQA with more than 42k questions and 556k evidences. As existing neural QA methods resolve QA either as sequence generation or classification/ranking problem, they face challenges of expensive softmax computation, unseen answers handling or separate candidate answer generation component. In this work, we cast neural QA as a sequence labeling problem and propose an end-to-end sequence labeling model, which overcomes all the above challenges. Experimental results on WebQA show that our model outperforms the baselines significantly with an F1 score of 74.69% with word-based input, and the performance drops only 3.72 F1 points with more challenging character-based input.

cs.CL↗

arXiv2016

A Data-Driven Approach for Mapping Multivariate Data to Color

A wide variety of color schemes have been devised for mapping scalar data to color. Some use the data value to index a color scale. Others assign colors to different, usually blended disjoint materials, to handle areas where materials overlap. A number of methods can map low-dimensional data to color, however, these methods do not scale to higher dimensional data. Likewise, schemes that take a more artistic approach through color mixing and the like also face limits when it comes to the number of variables they can encode. We address the challenge of mapping multivariate data to color and avoid these limitations at the same time. It is a data driven method, which first gauges the similarity of the attributes and then arranges them according to the periphery of a convex 2D color space, such as HSL. The color of a multivariate data sample is then obtained via generalized barycentric coordinate (GBC) interpolation.

cs.GR↗

arXiv2016

Deep Recurrent Models with Fast-Forward Connections for Neural Machine Translation

Neural machine translation (NMT) aims at solving machine translation (MT) problems using neural networks and has exhibited promising results in recent years. However, most of the existing NMT models are shallow and there is still a performance gap between a single NMT model and the best conventional MT system. In this work, we introduce a new type of linear connections, named fast-forward connections, based on deep Long Short-Term Memory (LSTM) networks, and an interleaved bi-directional architecture for stacking the LSTM layers. Fast-forward connections play an essential role in propagating the gradients and building a deep topology of depth 16. On the WMT'14 English-to-French task, we achieve BLEU=37.7 with a single attention model, which outperforms the corresponding single shallow model by 6.2 BLEU points. This is the first time that a single NMT model achieves state-of-the-art performance and outperforms the best conventional model by 0.7 BLEU points. We can still achieve BLEU=36.3 even without using an attention mechanism. After special handling of unknown words and model ensembling, we obtain the best score reported to date on this task with BLEU=40.4. Our models are also validated on the more difficult WMT'14 English-to-German task.

cs.CL↗

arXiv2016

CFO: Conditional Focused Neural Question Answering with Large-scale Knowledge Bases

How can we enable computers to automatically answer questions like "Who created the character Harry Potter"? Carefully built knowledge bases provide rich sources of facts. However, it remains a challenge to answer factoid questions raised in natural language due to numerous expressions of one question. In particular, we focus on the most common questions --- ones that can be answered with a single fact in the knowledge base. We propose CFO, a Conditional Focused neural-network-based approach to answering factoid questions with knowledge bases. Our approach first zooms in a question to find more probable candidate subject mentions, and infers the final answers with a unified conditional probabilistic framework. Powered by deep recurrent neural networks and neural embeddings, our proposed CFO achieves an accuracy of 75.7% on a dataset of 108k questions - the largest public one to date. It outperforms the current state of the art by an absolute margin of 11.8%.

cs.CL↗

arXiv2016

Debugging OpenStack Problems Using a State Graph Approach

It is hard to operate and debug systems like OpenStack that integrate many independently developed modules with multiple levels of abstractions. A major challenge is to navigate through the complex dependencies and relationships of the states in different modules or subsystems, to ensure the correctness and consistency of these states. We present a system that captures the runtime states and events from the entire OpenStack-Ceph stack, and automatically organizes these data into a graph that we call system operation state graph (SOSG).With SOSG we can use intuitive graph traversal techniques to solve problems like reasoning about the state of a virtual machine. Also, using graph-based anomaly detection, we can automatically discover hidden problems in OpenStack. We have a scalable implementation of SOSG, and evaluate the approach on a 125-node production OpenStack cluster, finding a number of interesting problems.

cs.DC↗

arXiv2016

Attention to Scale: Scale-aware Semantic Image Segmentation

Incorporating multi-scale features in fully convolutional neural networks (FCNs) has been a key element to achieving state-of-the-art performance on semantic image segmentation. One common way to extract multi-scale features is to feed multiple resized input images to a shared deep network and then merge the resulting features for pixelwise classification. In this work, we propose an attention mechanism that learns to softly weight the multi-scale features at each pixel location. We adapt a state-of-the-art semantic image segmentation model, which we jointly train with multi-scale input images and the attention model. The proposed attention model not only outperforms average- and max-pooling, but allows us to diagnostically visualize the importance of features at different positions and scales. Moreover, we show that adding extra supervision to the output at each scale is essential to achieving excellent performance when merging multi-scale features. We demonstrate the effectiveness of our model with extensive experiments on three challenging datasets, including PASCAL-Person-Part, PASCAL VOC 2012 and a subset of MS-COCO 2014.

cs.CV↗

arXiv2016

Multipair Massive MIMO Relaying with Pilot-Data Transmission Overlay

We propose a pilot-data transmission overlay scheme for multipair massive multiple-input multiple-output (MIMO) relaying systems employing either half- or full-duplex (HD or FD) communications at the relay station (RS). In the proposed scheme, pilots are transmitted in partial overlap with data to decrease the channel estimation overhead. The RS can detect the source data with minimal destination pilot interference by exploiting the asymptotic orthogonality of massive MIMO channels. Then pilot-data interference can be effectively suppressed with assistance of the detected source data in the destination channel estimation. Due to the transmission overlay, the effective data period is extended, hence improving system throughput. Both theoretical and simulation results confirm that the proposed pilot-data overlay scheme outperforms the conventional separate pilot-data design in the limited coherence time interval scenario. Moreover, asymptotic analyses at high and low SNR regions demonstrate the superiority of the proposed scheme regardless of the coherence interval length. Because of simultaneous transmission, the proper allocation of source data transmission and relay data forwarding power can further improve the system performance. Hence a power allocation problem is formulated and a successive convex approximation approach is proposed to solve the non-convex optimization problem with the FD pilot-data transmission overlay.

cs.IT↗

arXiv2016

Universal entropy relations: entropy formulae and entropy bound

We survey the applications of universal entropy relations in black holes with multi-horizons. In sharp distinction to conventional entropy product, the entropy relationship here not only improve our understanding of black hole entropy but was introduced as an elegant technique trick for handling various entropy bounds and sum. Despite the primarily technique role, entropy relations have provided considerable insight into several different types of gravity, including massive gravity, Einstein-Dilaton gravity and Horava-Lifshitz gravity. We present and discuss the results for each one.

gr-qc↗

arXiv2016

Study of Magnetic Hysteresis Effects in a Storage Ring Using Precision Tune Measurement

With advances in accelerator science and technology in the recent decades, the accelerator community has focused on the development of next-generation light sources, for example the diffraction-limited storage rings (DLSRs), which requires precision control of the electron beam energy and betatron tunes. This work is aimed at understanding magnet hysteresis effects on the electron beam energy and lattice focusing in the circular accelerators, and developing new methods to gain better control of these effects. In this paper, we will report our recent experimental study of the magnetic hysteresis effects and their impacts on the Duke storage ring lattice using the transverse feedback based precision tune measurement system. The major magnet hysteresis effects associated with magnet normalization and lattice ramping are carefully studied to determine an effective procedure for lattice preparation while maintaining a high degree of reproducibility of lattice focusing. The local hysteresis effects are also studied by measuring the betatron tune shifts resulted from adjusting the setting of a quadrupole. A new technique has been developed to precisely recover the focusing strength of the quadrupole by returning it to a proper setting to overcome the local hysteresis effect.

physics.acc-ph↗

arXiv2016

CNN-RNN: A Unified Framework for Multi-label Image Classification

While deep convolutional neural networks (CNNs) have shown a great success in single-label image classification, it is important to note that real world images generally contain multiple labels, which could correspond to different objects, scenes, actions and attributes in an image. Traditional approaches to multi-label image classification learn independent classifiers for each category and employ ranking or thresholding on the classification results. These techniques, although working well, fail to explicitly exploit the label dependencies in an image. In this paper, we utilize recurrent neural networks (RNNs) to address this problem. Combined with CNNs, the proposed CNN-RNN framework learns a joint image-label embedding to characterize the semantic label dependency as well as the image-label relevance, and it can be trained end-to-end from scratch to integrate both information in a unified framework. Experimental results on public benchmark datasets demonstrate that the proposed architecture achieves better performance than the state-of-the-art multi-label classification model

cs.CV↗

arXiv2016

Video Paragraph Captioning Using Hierarchical Recurrent Neural Networks

We present an approach that exploits hierarchical Recurrent Neural Networks (RNNs) to tackle the video captioning problem, i.e., generating one or multiple sentences to describe a realistic video. Our hierarchical framework contains a sentence generator and a paragraph generator. The sentence generator produces one simple short sentence that describes a specific short video interval. It exploits both temporal- and spatial-attention mechanisms to selectively focus on visual elements during generation. The paragraph generator captures the inter-sentence dependency by taking as input the sentential embedding produced by the sentence generator, combining it with the paragraph history, and outputting the new initial state for the sentence generator. We evaluate our approach on two large-scale benchmark datasets: YouTubeClips and TACoS-MultiLevel. The experiments demonstrate that our approach significantly outperforms the current state-of-the-art methods with BLEU@4 scores 0.499 and 0.305 respectively.

cs.CV↗

arXiv2016

ABC-CNN: An Attention Based Convolutional Neural Network for Visual Question Answering

We propose a novel attention based deep learning architecture for visual question answering task (VQA). Given an image and an image related natural language question, VQA generates the natural language answer for the question. Generating the correct answers requires the model's attention to focus on the regions corresponding to the question, because different questions inquire about the attributes of different image regions. We introduce an attention based configurable convolutional neural network (ABC-CNN) to learn such question-guided attention. ABC-CNN determines an attention map for an image-question pair by convolving the image feature map with configurable convolutional kernels derived from the question's semantics. We evaluate the ABC-CNN architecture on three benchmark VQA datasets: Toronto COCO-QA, DAQUAR, and VQA dataset. ABC-CNN model achieves significant improvements over state-of-the-art methods on these datasets. The question-guided attention generated by ABC-CNN is also shown to reflect the regions that are highly relevant to the questions.

cs.CV↗

arXiv2016

Nonparametric Estimation for Jump-Diffusion CIR Model

We study the nonparametric estimation for the intensity of Poisson random measure in jump-diffusion CIR model based on the low frequency observations. This is given in terms of the minimization of norms on a nonempty, closed and convex subset of some special Hilbert space. We establish the measurability of the estimator and derive its consistency and asymptotic risk bound.

math.ST↗