arXiv subjects

Xin Zhang

Publications and source records attributed to Xin Zhang.

At least 937 records · Page 52Linked to original sources

arXiv2021

Single pixel wide gamut dynamic color modulation based on graphene micromechanical system

Dynamic color modulation in the composite structure of graphene microelectromechanical systems (MEMS)- photonic crystal microcavity is investigated in this work. The designed photonic crystal microcavity has three resonant standing wave modes corresponding to the three primary colors of red (R), green (G) and blue (B), forming strong localization of light in three modes at different positions of the microcavity. Once graphene is added, it can govern the transmittance of three modes. When graphene is located in the abdomen of the standing wave, which has strong light absorption and therefore the structure's transmittance is lower, or when graphene is located in the node of the standing wave, it has weak light absorption and therefore the structure's transmittance is higher. Therefore, the graphene absorption of different colors of light can be regulated dynamically by applying voltages to tune the equilibrium position of the graphene MEMS in the microcavity, consequently realizing the output of vivid monochromatic light or multiple mixed colors of light within a single pixel, thus greatly improving the resolution. Our work provides a route to dynamic color modulation with graphene and provides guidance for the design and manufacture of ultrahigh resolution, ultrafast modulation and wide color gamut interferometric modulator displays.

physics.optics↗

arXiv2021

Towards interpreting computer vision based on transformation invariant optimization

Interpreting how does deep neural networks (DNNs) make predictions is a vital field in artificial intelligence, which hinders wide applications of DNNs. Visualization of learned representations helps we humans understand the vision of DNNs. In this work, visualized images that can activate the neural network to the target classes are generated by back-propagation method. Here, rotation and scaling operations are applied to introduce the transformation invariance in the image generating process, which we find a significant improvement on visualization effect. Finally, we show some cases that such method can help us to gain insight into neural networks.

cs.CV↗

arXiv2021

Forecast for cosmological parameter estimation with gravitational-wave standard sirens from the LISA-Taiji network

LISA and Taiji are expected to form a space-based gravitational-wave (GW) detection network in the future. In this work, we make a forecast for the cosmological parameter estimation with the standard siren observation from the LISA-Taiji network. We simulate the standard siren data based on a scenario with configuration angle of $40^{\circ}$ between LISA and Taiji. Three models for the population of massive black hole binary (MBHB), i.e., pop III, Q3d, and Q3nod, are considered to predict the events of MBHB mergers. We find that, based on the LISA-Taiji network, the number of electromagnetic (EM) counterparts detected is almost doubled compared with the case of single Taiji mission. Therefore, the LISA-Taiji network's standard siren observation could provide much tighter constraints on cosmological parameters. For example, solely using the standard sirens from the LISA-Taiji network, the constraint precision of $H_0$ could reach $1.3\%$. Moreover, combined with the CMB data, the GW-EM observation based on the LISA-Taiji network could also tightly constrain the equation of state of dark energy, e.g., the constraint precision of $w$ reaches about $4\%$, which is comparable with the result of CMB+BAO+SN. It is concluded that the GW standard sirens from the LISA-Taiji network will become a useful cosmological probe in understanding the nature of dark energy in the future.

gr-qc↗

arXiv2021

Nuclear reaction measurements of 80.5 MeV/u 12C beam bombarding on C, W, Cu, Au, Pb targets

To get the energy spectrum distribution and cross-sections of emitted light charged particles and explore the nuclear reaction, a experiment of 80.5 MeV/u 12C beam bombarding on C, W, Cu, Au, Pb targets has been carried out at Institute of Modern Physics, Chinese Academy of Science. 30, 60 and 120 degree relative to the incident beam have been detected using three sets of telescope detectors. The results indicate that there is a tendency that heavier targets have larger double differential cross-sections and the emitted fragments are more likely to go forward. Besides, the decrease of cross-sections of fragments producing with the increasing emitted angles may follow some kind of pattern.

nucl-ex↗

arXiv2021

Relieving the $H_0$ tension with a new interacting dark energy model

We investigate an extended cosmological model motivated by the asymptotic safety of gravitational field theory, in which the matter and radiation densities and the cosmological constant receive a correction parametrized by the parameters $δ_G$ and $δ_Λ$, leading to that both the evolutions of the matter and radiation densities and the cosmological constant slightly deviate from the standard forms. Here we explain this model as a scenario of vacuum energy interacting with matter and radiation. We consider two cases of the model: {(i) ${\tildeΛ}$CDM with one additional free parameter $δ_G$, with $δ_{\rm G}$ and $δ_Λ$ related by a low-redshift limit relation and (ii) e${\tildeΛ}$CDM with two additional free parameters $δ_G$ and $δ_Λ$ that are independent of each other.} We use two data combinations, CMB+BAO+SN (CBS) and CMB+BAO+SN+$H_0$ (CBSH), to constrain the models. We find that, in the case of using the CBS data, neither ${\tildeΛ}$CDM nor e${\tildeΛ}$CDM can effectively alleviate the $H_0$ tension. However, it is found that using the CBSH data the $H_0$ tension can be greatly relieved by the models. In particular, in the case of e${\tildeΛ}$CDM, the $H_0$ tension can be resolved to 0.71$σ$. We conclude that as an interacting dark energy model, ${\tildeΛ}$CDM is much better than $Λ(t)$CDM in the sense of both relieving the $H_0$ tension and fitting to the current observational data.

astro-ph.CO↗

arXiv2021

Negligible oxygen vacancies, low critical current density, electric-field modulation, in-plane anisotropic and high-field transport of a superconducting Nd0.8Sr0.2NiO2/SrTiO3 heterostructure

The emerging Ni-based superconducting oxide thin films are rather intriguing to the entire condensed matter physics. Here we report some brief experimental results on transport measurements for a 14-nm-thick superconducting Nd0.8Sr0.2NiO2/SrTiO3 thin-film heterostructure with an onset transition temperature of ~9.5 K. Photoluminescence measurements reveal that there is negligible oxygen vacancy creation in the SrTiO3 substrate during thin-film deposition and post chemical reduction for the Nd0.8Sr0.2NiO2/SrTiO3 heterostructure. It was found that the critical current density of the Nd0.8Sr0.2NiO2/SrTiO3 thin-film heterostructure is relatively small, ~4x10^3 A/cm2. Although the surface steps of SrTiO3 substrates lead to an anisotropy for in-plane resistivity, the superconducting transition temperatures are almost the same. The out-of-plane magnetotransport measurements yield an upper critical field of ~11.4 T and an estimated in-plane Ginzburg-Landau coherence length of ~5.4 nm. High-field magnetotransport measurements up to 50 T reveal anisotropic critical fields at 1.8 K for three different measurement geometries and a complicated Hall effect. An electric field applied via the SrTiO3 substrate slightly varies the superconducting transition temperature. These experimental results could be useful for this rapidly developing field.

cond-mat.supr-con↗

arXiv2021

Glauber Gluon Effects in Soft Collinear Factorization

Effects of Glauber gluons, which cause the elastic scattering process between different jets, are studied in the frame of soft-collinear effective theory(SCET). Glauber modes are added into the Lagrangian before integrated out, which is helpful in studies on Glauber couplings of collinear and soft particles explicitly. It is proved that interactions after the hard collision cancel out in processes inclusive enough. So are interactions with the light cone coordinates Sx^{+}$ and $x^{-}$ greater than those of the hard collision. Eikonalization of Glauber couplings of active particles and absorption of active-active and active-soft and active-spectator Glauber exchanges into soft and collinear Wilson lines are discussed, which is related to loop level definitions of Glauber gluons here. The active-spectator coherence are proved to be harmless on inclusive summation of spectator-spectator and spectator-soft Glauber exchanges. Based on this result, spectator-spectator and spectator-soft Glauber exchanges are proved to cancel out in processes considered here. Graphic aspects of the cancellation are also discussed to explain relations between the graphic cancellation of Glauber gluons in the frame of perturbative QCD and operator level skills in the paper.

hep-ph↗

arXiv2021

GT-STORM: Taming Sample, Communication, and Memory Complexities in Decentralized Non-Convex Learning

Decentralized nonconvex optimization has received increasing attention in recent years in machine learning due to its advantages in system robustness, data privacy, and implementation simplicity. However, three fundamental challenges in designing decentralized optimization algorithms are how to reduce their sample, communication, and memory complexities. In this paper, we propose a \underline{g}radient-\underline{t}racking-based \underline{sto}chastic \underline{r}ecursive \underline{m}omentum (GT-STORM) algorithm for efficiently solving nonconvex optimization problems. We show that to reach an $ε^2$-stationary solution, the total number of sample evaluations of our algorithm is $\tilde{O}(m^{1/2}ε^{-3})$ and the number of communication rounds is $\tilde{O}(m^{-1/2}ε^{-3})$, which improve the $O(ε^{-4})$ costs of sample evaluations and communications for the existing decentralized stochastic gradient algorithms. We conduct extensive experiments with a variety of learning models, including non-convex logistical regression and convolutional neural networks, to verify our theoretical findings. Collectively, our results contribute to the state of the art of theories and algorithms for decentralized network optimization.

cs.LG↗

arXiv2021

Anisotropic g-Factor and Spin-Orbit Field in a Ge Hut Wire Double Quantum Dot

Holes in nanowires have drawn significant attention in recent years because of the strong spin-orbit interaction, which plays an important role in constructing Majorana zero modes and manipulating spin-orbit qubits. Here, from the strongly anisotropic leakage current in the spin blockade regime for a double dot, we extract the full g-tensor and find that the spin-orbit field is in plane with an azimuthal angle of 59° to the axis of the nanowire. The direction of the spin-orbit field indicates a strong spin-orbit interaction along the nanowire, which may have originated from the interface inversion asymmetry in Ge hut wires. We also demonstrate two different spin relaxation mechanisms for the holes in the Ge hut wire double dot: spin-flip cotunneling to the leads, and spin-orbit interaction within the double dot. These results help establish feasibility of a Ge-based quantum processor.

cond-mat.mes-hall↗

arXiv2021

Controlling Synthetic Spin-Orbit Coupling in a Silicon Quantum Dot with Magnetic Field

Tunable synthetic spin-orbit coupling (s-SOC) is one of the key challenges in various quantum systems, such as ultracold atomic gases, topological superconductors, and semiconductor quantum dots. Here we experimentally demonstrate controlling the s-SOC by investigating the anisotropy of spin-valley resonance in a silicon quantum dot. As we rotate the applied magnetic field in-plane, we find a striking nonsinusoidal behavior of resonance amplitude that distinguishes s-SOC from the intrinsic spin-orbit coupling (i-SOC), and associate this behavior with the previously overlooked in-plane transverse magnetic field gradient. Moreover, by theoretically analyzing the experimentally measured s-SOC field, we predict the quality factor of the spin qubit could be optimized if the orientation of the in-plane magnetic field is rotated away from the traditional working point.

cond-mat.mes-hall↗

arXiv2021

Cross-Modality Brain Tumor Segmentation via Bidirectional Global-to-Local Unsupervised Domain Adaptation

Accurate segmentation of brain tumors from multi-modal Magnetic Resonance (MR) images is essential in brain tumor diagnosis and treatment. However, due to the existence of domain shifts among different modalities, the performance of networks decreases dramatically when training on one modality and performing on another, e.g., train on T1 image while performing on T2 image, which is often required in clinical applications. This also prohibits a network from being trained on labeled data and then transferred to unlabeled data from a different domain. To overcome this, unsupervised domain adaptation (UDA) methods provide effective solutions to alleviate the domain shift between labeled source data and unlabeled target data. In this paper, we propose a novel Bidirectional Global-to-Local (BiGL) adaptation framework under a UDA scheme. Specifically, a bidirectional image synthesis and segmentation module is proposed to segment the brain tumor using the intermediate data distributions generated for the two domains, which includes an image-to-image translator and a shared-weighted segmentation network. Further, a global-to-local consistency learning module is proposed to build robust representation alignments in an integrated way. Extensive experiments on a multi-modal brain MR benchmark dataset demonstrate that the proposed method outperforms several state-of-the-art unsupervised domain adaptation methods by a large margin, while a comprehensive ablation study validates the effectiveness of each key component. The implementation code of our method will be released at \url{https://github.com/KeleiHe/BiGL}.

cs.CV↗

arXiv2021

Constructing long-short stock portfolio with a new listwise learn-to-rank algorithm

Factor strategies have gained growing popularity in industry with the fast development of machine learning. Usually, multi-factors are fed to an algorithm for some cross-sectional return predictions, which are further used to construct a long-short portfolio. Instead of predicting the value of the stock return, emerging studies predict a ranked stock list using the mature learn-to-rank technology. In this study, we propose a new listwise learn-to-rank loss function which aims to emphasize both the top and the bottom of a rank list. Our loss function, motivated by the long-short strategy, is endogenously shift-invariant and can be viewed as a direct generalization of ListMLE. Under different transformation functions, our loss can lead to consistency with binary classification loss or permutation level 0-1 loss. A probabilistic explanation for our model is also given as a generalized Plackett-Luce model. Based on a dataset of 68 factors in China A-share market from 2006 to 2019, our empirical study has demonstrated the strength of our method which achieves an out-of-sample annual return of 38% with the Sharpe ratio being 2.

q-fin.PM↗

arXiv2021

Generalized Liquid Association Analysis for Multimodal Data Integration

Multimodal data are now prevailing in scientific research. A central question in multimodal integrative analysis is to understand how two data modalities associate and interact with each other given another modality or demographic variables. The problem can be formulated as studying the associations among three sets of random variables, a question that has received relatively less attention in the literature. In this article, we propose a novel generalized liquid association analysis method, which offers a new and unique angle to this important class of problems of studying three-way associations. We extend the notion of liquid association of \citet{li2002LA} from the univariate setting to the sparse, multivariate, and high-dimensional setting. We establish a population dimension reduction model, transform the problem to sparse Tucker decomposition of a three-way tensor, and develop a higher-order orthogonal iteration algorithm for parameter estimation. We derive the non-asymptotic error bound and asymptotic consistency of the proposed estimator, while allowing the variable dimensions to be larger than and diverge with the sample size. We demonstrate the efficacy of the method through both simulations and a multimodal neuroimaging application for Alzheimer's disease research.

stat.ME↗

arXiv2021

A Doubly-Enhanced EM Algorithm for Model-Based Tensor Clustering

Modern scientific studies often collect data sets in the forms of tensors, which call for innovative statistical analysis methods. In particular, there is a pressing need for tensor clustering methods to understand the heterogeneity in the data. We propose a tensor normal mixture model (TNMM) approach to enable probabilistic interpretation and computational tractability. Our statistical model leverages the tensor covariance structure to reduce the number of parameters for parsimonious modeling, and at the same time explicitly exploits the correlations for better variable selection and clustering. We propose a doubly-enhanced expectation-maximization (DEEM) algorithm to perform clustering under this model. Both the E-step and the M-step are carefully tailored for tensor data in order to account for statistical accuracy and computational cost in high dimensions. Theoretical studies confirm that DEEM achieves consistent clustering even when the dimension of each mode of the tensors grows at an exponential rate of the sample size. Numerical studies demonstrate favorable performance of DEEM in comparison to existing methods.

stat.ME↗

arXiv2021

An Explainable 3D Residual Self-Attention Deep Neural Network FOR Joint Atrophy Localization and Alzheimer's Disease Diagnosis using Structural MRI

Computer-aided early diagnosis of Alzheimer's disease (AD) and its prodromal form mild cognitive impairment (MCI) based on structure Magnetic Resonance Imaging (sMRI) has provided a cost-effective and objective way for early prevention and treatment of disease progression, leading to improved patient care. In this work, we have proposed a novel computer-aided approach for early diagnosis of AD by introducing an explainable 3D Residual Attention Deep Neural Network (3D ResAttNet) for end-to-end learning from sMRI scans. Different from the existing approaches, the novelty of our approach is three-fold: 1) A Residual Self-Attention Deep Neural Network has been proposed to capture local, global and spatial information of MR images to improve diagnostic performance; 2) An explanation method using Gradient-based Localization Class Activation mapping (Grad-CAM) has been introduced to improve the explainable of the proposed method; 3) This work has provided a full end-to-end learning solution for automated disease diagnosis. Our proposed 3D ResAttNet method has been evaluated on a large cohort of subjects from real datasets for two changeling classification tasks (i.e., Alzheimer's disease (AD) vs. Normal cohort (NC) and progressive MCI (pMCI) vs. stable MCI (sMCI)). The experimental results show that the proposed approach has a competitive advantage over the state-of-the-art models in terms of accuracy performance and generalizability. The explainable mechanism in our approach is able to identify and highlight the contribution of the important brain parts (e.g., hippocampus, lateral ventricle and most parts of the cortex) for transparent decisions.

eess.IV↗

arXiv2021

Equitable vertex arboricity conjecture holds for graphs with low degeneracy

The equitable tree-coloring can formulate a structure decomposition problem on the communication network with some security considerations. Namely, an equitable tree-$k$-coloring of a graph is a vertex coloring using $k$ distinct colors such that every color class induces a forest and the sizes of any two color classes differ by at most one. In this paper, we show some theoretical results on the equitable tree-coloring of graphs by proving that every $d$-degenerate graph with maximum degree at most $Δ$ is equitably tree-$k$-colorable for every integer $k\geq (Δ+1)/2$ provided that $Δ\geq 9.818d$, confirming the equitable vertex arboricity conjecture for graphs with low degeneracy.

math.CO↗

arXiv2021

Bayesian Full-waveform Inversion with Realistic Priors

Seismic full-waveform inversion (FWI) uses full seismic records to estimate subsurface velocity structure. This requires a highly nonlinear and nonunique inverse problem to be solved, and Bayesian methods have been used to quantify uncertainties in the solution. Variational Bayesian inference uses optimization to provide solutions efficiently. The method has been applied to solve a transmission FWI problem using data generated by known earthquake-like sources, with strong prior information imposed on the velocity. Unfortunately such prior information about velocity structure and earthquake sources is never available in practice. We present the first application of the method in a seismic reflection setting, and with realistically weak prior information. We thus demonstrate that the method can produce high resolution images and reliable uncertainties given practically reasonable prior information.

physics.geo-ph↗

arXiv2021

Power-aligned 2HDM: a correlative perspective on $(g-2)_{e,μ}$

With the hypothesis of minimal flavor violation, we find that there exists a power-aligned relation between the Yukawa couplings of the two scalar doublets in the two-Higgs-doublet model with Hermitian Yukawa matrices. Within such a power-aligned framework, it is found that a simultaneous explanation of the anomalies observed in the electron and muon anomalous magnetic moments can be reached with TeV-scale quasi-degenerate Higgs masses, and the resulting parameter space is also phenomenologically safer under the B-physics, $Z$ and $τ$ decay data, as well as the current LHC bounds. Furthermore, the flavor-universal power that enhances the charged-lepton Yukawa couplings prompts an interesting correlation between the two anomalies, which makes the model distinguishable from the (generalized) linearly aligned and the lepton-specific two-Higgs-doublet models that address the same anomalies but in a non-correlative manner, and hence testable by future precise measurements.

hep-ph↗