arXiv subjects

Yong Zhao

Publications and source records attributed to Yong Zhao.

At least 217 records · Page 12Linked to original sources

arXiv2019

GSLAM: A General SLAM Framework and Benchmark

SLAM technology has recently seen many successes and attracted the attention of high-technological companies. However, how to unify the interface of existing or emerging algorithms, and effectively perform benchmark about the speed, robustness and portability are still problems. In this paper, we propose a novel SLAM platform named GSLAM, which not only provides evaluation functionality, but also supplies useful toolkit for researchers to quickly develop their own SLAM systems. The core contribution of GSLAM is an universal, cross-platform and full open-source SLAM interface for both research and commercial usage, which is aimed to handle interactions with input dataset, SLAM implementation, visualization and applications in an unified framework. Through this platform, users can implement their own functions for better performance with plugin form and further boost the application to practical usage of the SLAM.

cs.CV↗

arXiv2019

Determining the Nonperturbative Collins-Soper Kernel From Lattice QCD

At small transverse momentum $q_T$, transverse-momentum dependent parton distribution functions (TMDPDFs) arise as genuinely nonperturbative objects that describe Drell-Yan like processes in hadron collisions as well as semi-inclusive deep-inelastic scattering. TMDPDFs naturally depend on the hadron momentum, and the associated evolution is determined by the Collins-Soper equation. For $q_T \sim Λ_\mathrm{QCD}$ the corresponding evolution kernel (or anomalous dimension) is nonperturbative and must be determined as an independent ingredient in order to relate TMDPDFs at different scales. We propose a method to extract this kernel using lattice QCD and the Large-Momentum Effective Theory, where the physical TMD correlation involving light-like paths is approximated by a quasi TMDPDF, defined using equal-time correlation functions with a large-momentum hadron state. The kernel is determined from a ratio of quasi TMDPDFs extracted at different hadron momenta.

hep-ph↗

arXiv2019

Proton Isovector Helicity Distribution on the Lattice at Physical Pion Mass

We present a state-of-the-art calculation of the isovector quark helicity Bjorken-$x$ distribution in the proton using lattice-QCD ensembles at the physical pion mass. We compute quasi-distributions at proton momenta $P_z \in \{2.2, 2.6, 3.0\}$~GeV on the lattice, and match them systematically to the physical parton distribution using large-momentum effective theory (LaMET). We reach an unprecedented precision through high statistics in simulations, large-momentum proton matrix elements, and control of excited-state contamination. The resulting distribution with combined statistical and systematic errors is in agreement with the latest phenomenological analysis of the spin-dependent experimental data; in particular, $Δ\bar{u}(x)>Δ\bar{d}(x)$.

hep-lat↗

arXiv2019

Speaker Adaptation for End-to-End CTC Models

We propose two approaches for speaker adaptation in end-to-end (E2E) automatic speech recognition systems. One is Kullback-Leibler divergence (KLD) regularization and the other is multi-task learning (MTL). Both approaches aim to address the data sparsity especially output target sparsity issue of speaker adaptation in E2E systems. The KLD regularization adapts a model by forcing the output distribution from the adapted model to be close to the unadapted one. The MTL utilizes a jointly trained auxiliary task to improve the performance of the main task. We investigated our approaches on E2E connectionist temporal classification (CTC) models with three different types of output units. Experiments on the Microsoft short message dictation task demonstrated that MTL outperforms KLD regularization. In particular, the MTL adaptation obtained 8.8\% and 4.0\% relative word error rate reductions (WERRs) for supervised and unsupervised adaptations for the word CTC model, and 9.6% and 3.8% relative WERRs for the mix-unit CTC model, respectively.

cs.CL↗

arXiv2018

Moments of pion distribution amplitude using operator product expansion on the lattice

We report an exploratory study of the current-current matrix elements that are relevant to the extraction of moments of the pion light-cone distribution amplitude, employing the method of introducing a valence relativistic heavy quark. The numerical investigation is carried out in the quenched approximation with the physical volume $L\approx 2.4$ fm at two values of lattice spacing (0.05 and 0.075 fm). We obtain clean signals for the relevant Euclidean hadronic tensor with reasonable statistics, but observe that the lattice artefacts are non-negligible in our results. The key conclusion from the analysis hitherto is that although our approach has the potential for making significant contributions to parton physics, data at finer lattice spacings that are currently being produced are needed in order to control the continuum extrapolation.

hep-lat↗

arXiv2018

Matching the Quasi Meson Distribution Amplitude in RI/MOM scheme

The $x$-dependence of light-cone distribution amplitude (LCDA) can be directly calculated from a quasi distribution amplitude (DA) in lattice QCD within the framework of large-momentum effective theory (LaMET). In this paper, we study the one-loop renormalization of the quasi-DA in the regularization-independent momentum subtraction (RI/MOM) scheme. The renormalization factor for the quasi parton distribution function can be used to renormalize the quasi-DA provided that they are implemented on lattice and in perturbation theory in the same manner. We derive the one-loop matching coefficient that matches quasi-DA in the RI/MOM scheme onto LCDA in the $\overline{\rm MS}$ scheme. Our result provides the crucial step to extract the LCDAs from lattice matrix elements of quasi-DAs.

hep-ph↗

arXiv2018

Nucleon Transversity Distribution at the Physical Pion Mass from Lattice QCD

We report a state-of-the-art lattice calculation of the isovector quark transversity distribution of the proton at the physical pion mass. Within the framework of large-momentum effective theory (LaMET), we compute the transversity quasi-distributions using clover valence fermions on 2+1+1-flavor (up/down, strange, charm) HISQ-lattice configurations with boosted proton momenta as large as 3.0~GeV. The relevant lattice matrix elements are nonperturbatively renormalized in regularization-independent momentum-subtraction (RI/MOM) scheme and systematically matched to the physical transversity distribution. With high statistics, large proton momenta and meticulous control of excited-state contamination, we provide the best theoretical prediction for the large-$x$ isovector quark transversity distribution, with better precision than the most recent global analyses of experimental data. Our result also shows that the sea quark asymmetry in the proton transversity distribution is consistent with zero, which has been assumed in all current global analyses.

hep-lat↗

arXiv2018

Factorization Theorem Relating Euclidean and Light-Cone Parton Distributions

In a large-momentum nucleon state, the matrix element of a gauge-invariant Euclidean Wilson line operator accessible from lattice QCD can be related to the standard light-cone parton distribution function through the large-momentum effective theory (LaMET) expansion. This relation is given by a factorization theorem with a non-trivial matching coefficient. Using the operator product expansion we prove the large-momentum factorization of the quasi-parton distribution function in LaMET, and show that the more recently discussed Ioffe-time distribution approach also obeys an equivalent factorization theorem. Explicit results for the coefficients are obtained and compared at one-loop. Our proof clearly demonstrates that the matching coefficients in the $\overline{\rm MS}$ scheme depend on the large partonic momentum rather than the nucleon momentum.

hep-ph↗

arXiv2018

Lattice Calculation of Parton Distribution Function from LaMET at Physical Pion Mass with Large Nucleon Momentum

We present a lattice-QCD calculation of the unpolarized isovector parton distribution function (PDF) using ensembles at the physical pion mass with large proton boost momenta $P_z \in \{2.2,2.6,3.0\}$~GeV within the framework of large-momentum effective theory (LaMET). In contrast to our previous physical-pion PDF result, we increase the statistics significantly, double the boost momentum, increase the investment in excited-state contamination systematics, and switch to $γ_t$ operator to avoid mixing with scalar matrix elements. We use four source-sink separations in our analysis to control the systematics associated with excited-state contamination. The one-loop LaMET matching corresponding to the new operator is calculated and applied to our lattice data. We detail the systematics that affect PDF calculations, providing guidelines to improve the precision of future lattice PDF calculations. We find our final parton distribution to be in reasonable agreement with the PDF provided by the latest phenomenological analysis.

hep-lat↗

arXiv2018

Investigating the Efficiency of the Beijing Faint Object Spectrograph and Camera (BFOSC) of the Xinglong 2.16-m Reflector

The Beijing Faint Object Spectrograph and Camera (BFOSC) is one of the most important instruments of the 2.16-m telescope of the Xinglong Observatory. Every year there are ~ 20 SCI-papers published based on the observational data of this telescope. In this work, we have systemically measured the total efficiency of the BFOSC of the 2.16-m reflector, based on the observations of two ESO flux standard stars. We have obtained the total efficiencies of the BFOSC instrument of different grisms with various slit widths in almost all ranges, and analysed the factors which effect the efficiency of telescope and spectrograph. For the astronomical observers, the result will be useful for them to select a suitable slit width, depending on their scientific goals and weather conditions during the observation; For the technicians, the result will help them systemically find out the real efficiency of telescope and spectrograph, and further to improve the total efficiency and observing capacity of the telescope technically.

astro-ph.IM↗

arXiv2018

Upgraded Photometric System of The 85-cm Telescope at Xinglong Station

The 85-cm telescope at the Xinglong station is a well-operated prime focus system with high science outputs. The telescope has been upgraded since 2014 with new corrector, filters and camera, which are provided by Beijing Normal University (BNU). The filter set is Johnson-Cousins UBVRI system. We report the test results of the new system including the bias, dark current, linearity, gain and readout noise of the CCD camera . Then we derive accurate instrumental calibration coefficients in UBVRI bands with Landolt standard stars in the photometric nights. Finally, we give the limiting magnitudes with various exposure time and signal-to-noise ratio for observers as references.

astro-ph.IM↗

arXiv2018

Cascaded multi-scale and multi-dimension convolutional neural network for stereo matching

Convolutional neural networks(CNN) have been shown to perform better than the conventional stereo algorithms for stereo estimation. Numerous efforts focus on the pixel-wise matching cost computation, which is the important building block for many start-of-the-art algorithms. However, those architectures are limited to small and single scale receptive fields and use traditional methods for cost aggregation or even ignore cost aggregation. Differently we take them both into consideration. Firstly, we propose a new multi-scale matching cost computation sub-network, in which two different sizes of receptive fields are implemented parallelly. In this way, the network can make the best use of both variants and balance the trade-off between the increase of receptive field and the loss of detail. Furthermore, we show that our multi-dimension aggregation sub-network which containing 2D convolution and 3D convolution operations can provide rich context and semantic information for estimating an accurate initial disparity. Finally, experiments on challenging stereo benchmark KITTI demonstrate that the proposed method can achieve competitive results even without any additional post-processing.

cs.CV↗

arXiv2018

Matching the Quasi Parton Distribution in a Momentum Subtraction Scheme

The quasi parton distribution is a spatial correlation of quarks or gluons along the $z$ direction in a moving nucleon which enables direct lattice calculations of parton distribution functions. It can be defined with a nonperturbative renormalization in a regularization independent momentum subtraction scheme (RI/MOM), which can then be perturbatively related to the collinear parton distribution in the $\overline{\text{MS}}$ scheme. Here we carry out a direct matching from the RI/MOM scheme for the quasi-PDF to the $\overline{\text{MS}}$ PDF, determining the non-singlet quark matching coefficient at next-to-leading order in perturbation theory. We find that the RI/MOM matching coefficient is insensitive to the ultraviolet region of convolution integral, exhibits improved perturbative convergence when converting between the quasi-PDF and PDF, and is consistent with a quasi-PDF that vanishes in the unphysical region as the proton momentum $P^z\to \infty$, unlike other schemes. This direct approach therefore has the potential to improve the accuracy for converting quasi-distribution lattice calculations to collinear distributions.

hep-ph↗

arXiv2018

Kaon Distribution Amplitude from Lattice QCD and the Flavor SU(3) Symmetry

We present the first lattice-QCD calculation of the kaon distribution amplitude using the large-momentum effective theory (LaMET) approach. The momentum-smearing technique has been implemented to improve signals at large meson momenta. We subtract the power divergence due to Wilson line to high precision using multiple lattice spacings. The kaon structure clearly shows an asymmetry of the distribution amplitude around $x=1/2$, a clear sign of its skewness. We also study the leading SU(3) flavor symmetry breaking relations for the pion, kaon and eta meson distribution amplitudes, and the results are consistent with the prediction from chiral perturbation theory.

hep-ph↗

arXiv2018

Integrating quantum key distribution with classical communications in backbone fiber network

Quantum key distribution (QKD) provides information-theoretic security based on the laws of quantum mechanics. The desire to reduce costs and increase robustness in real-world applications has motivated the study of coexistence between QKD and intense classical data traffic in a single fiber. Previous works on coexistence in metropolitan areas have used wavelength-division multiplexing, however, coexistence in backbone fiber networks remains a great experimental challenge, as Tbps data of up to 20 dBm optical power is transferred, and much more noise is generated for QKD. Here we present for the first time, to the best of our knowledge, the integration of QKD with a commercial backbone network of 3.6 Tbps classical data at 21 dBm launch power over 66 km fiber. With 20 GHz pass-band filtering and large effective core area fibers, real-time secure key rates can reach 4.5 kbps and 5.1 kbps for co-propagation and counter-propagation at the maximum launch power, respectively. This demonstrates feasibility and represents an important step towards building a quantum network that coexists with the current backbone fiber infrastructure of classical communications.

quant-ph↗

arXiv2018

Parton Distribution Function with Non-perturbative Renormalization from Lattice QCD

We present lattice results for the isovector unpolarized parton distribution with nonperturbative RI/MOM-scheme renormalization on the lattice. In the framework of large-momentum effective field theory (LaMET), the full Bjorken-$x$ dependence of a momentum-dependent quasi-distribution is calculated on the lattice and matched to the ordinary lightcone parton distribution at one-loop order, with power corrections included. The important step of RI/MOM renormalization that connects the lattice and continuum matrix elements is detailed in this paper. A few consequences of the results are also addressed here.

hep-lat↗

arXiv2018

Extend the shallow part of Single Shot MultiBox Detector via Convolutional Neural Network

Single Shot MultiBox Detector (SSD) is one of the fastest algorithms in the current object detection field, which uses fully convolutional neural network to detect all scaled objects in an image. Deconvolutional Single Shot Detector (DSSD) is an approach which introduces more context information by adding the deconvolution module to SSD. And the mean Average Precision (mAP) of DSSD on PASCAL VOC2007 is improved from SSD's 77.5% to 78.6%. Although DSSD obtains higher mAP than SSD by 1.1%, the frames per second (FPS) decreases from 46 to 11.8. In this paper, we propose a single stage end-to-end image detection model called ESSD to overcome this dilemma. Our solution to this problem is to cleverly extend better context information for the shallow layers of the best single stage (e.g. SSD) detectors. Experimental results show that our model can reach 79.4% mAP, which is higher than DSSD and SSD by 0.8 and 1.9 points respectively. Meanwhile, our testing speed is 25 FPS in Titan X GPU which is more than double the original DSSD.

cs.CV↗

arXiv2018

Energy Efficiency Analysis of Heterogeneous Cellular Networks With Extra Cell Range Expansion

The split control and user plane is key to the future heterogeneous cellular network (HCN), where the small cells are dedicated for the most data transmission while the macrocells are mainly responsible for the control signaling. Adapting to this technology, we propose a general and tractable framework of extra cell range expansion (CRE) by introducing an additional bias factor to enlarge the range of small cells flexibly for the extra offloaded macrousers in a two-tier HCN, where the macrocell and small cell users have different required data rates. Using stochastic geometry, we analyze the energy efficiency (EE) of the extra CRE with joint low power transmission and resource partitioning, where the coverages of EE and data rate are formulated theoretically. Numerical simulations verify that the proposed extra CRE can improve the EE performance of HCN, and also show that deploying more small cells can provide benefits for EE coverage, but the EE improvement becomes saturated if the small cell density exceeds a threshold. Instead of establishing the detail configuration, this paper can provide some valuable insights and guidelines to the practical design of future networks, especially for the traffic offloading in HCN.

cs.IT↗