Search arXiv⌕ Search

arXiv subjects

Yu Zhang

Publications and source records attributed to Yu Zhang.

At least 1,279 records · Page 71Linked to original sources
arXiv2021

Analytical time-like geodesics in modified Hayward black hole space-time

The properties of modified Hayward black hole space-time can be investigated through analyzing the particle geodesics. By means of a detailed analysis of the corresponding effective potentials for a massive particle, we find all possible orbits which are allowed by the energy levels. The trajectories of orbits are plotted by solving the equation of orbital motion numerically. We conclude that whether there is an escape orbit is associated with $b$ (angular momentum). The properties of orbital motion are related to $b$, $α$ ($α$ is associated with the time delay) and $β$ ($β$ is related to 1-loop quantum corrections). There are no escape orbits when $b$ $<$ $4.016M$, $α$ = 0.50 and $β$ = 1.00. For fixed $α$ = 0.50 and $β$ = 1.00, if $b$ $<$ $3.493M$, there only exist unstable orbits. Comparing with the regular Hayward black hole, we go for a reasonable speculation by mean of the existing calculating results that the introduction of the modified term makes the radius of the innermost circular orbit (ISCO) and the corresponding angular momentum larger.

gr-qc↗
arXiv2021

Structure of geodesics in the regular Hayward black hole space-time

The regular Hayward model describes a non-singular black hole space-time. By analyzing the behaviors of effective potential and solving the equation of orbital motion, we investigate the time-like and null geodesics in the regular Hayward black hole space-time. Through detailed analyses of corresponding effective potentials for massive particles and photons, all possible orbits are numerically simulated. The results show that there may exist four orbital types in the time-like geodesics structure: planetary orbits, circular orbits, escape orbits and absorbing orbits. In addition, when $\ell$, a convenient encoding of the central energy density $3/8π\ell^{2}$, is $0.6M$, and $b$ is $3.9512M$ as a specific value of angular momentum, escape orbits exist only under $b>3.9512M$. The precession direction is also associated with values of $b$. With $b=3.70M$ the bound orbits shift clockwise but counter-clockwise with $b=5.00M$ in the regular Hayward black hole space-time. We also find that the structure of null geodesics is simpler than that of time-like geodesics. There only exist three kinds of orbits (unstable circle orbits, escape orbits and absorbing orbits).

gr-qc↗
arXiv2021

FedSAE: A Novel Self-Adaptive Federated Learning Framework in Heterogeneous Systems

Federated Learning (FL) is a novel distributed machine learning which allows thousands of edge devices to train model locally without uploading data concentrically to the server. But since real federated settings are resource-constrained, FL is encountered with systems heterogeneity which causes a lot of stragglers directly and then leads to significantly accuracy reduction indirectly. To solve the problems caused by systems heterogeneity, we introduce a novel self-adaptive federated framework FedSAE which adjusts the training task of devices automatically and selects participants actively to alleviate the performance degradation. In this work, we 1) propose FedSAE which leverages the complete information of devices' historical training tasks to predict the affordable training workloads for each device. In this way, FedSAE can estimate the reliability of each device and self-adaptively adjust the amount of training load per client in each round. 2) combine our framework with Active Learning to self-adaptively select participants. Then the framework accelerates the convergence of the global model. In our framework, the server evaluates devices' value of training based on their training loss. Then the server selects those clients with bigger value for the global model to reduce communication overhead. The experimental result indicates that in a highly heterogeneous system, FedSAE converges faster than FedAvg, the vanilla FL framework. Furthermore, FedSAE outperforms than FedAvg on several federated datasets - FedSAE improves test accuracy by 26.7% and reduces stragglers by 90.3% on average.

cs.LG↗
arXiv2021

Fronthaul Compression and Passive Beamforming Design for Intelligent Reflecting Surface-aided Cloud Radio Access Networks

This letter studies a cloud radio access network (C-RAN) with multiple intelligent reflecting surfaces (IRS) deployed between users and remote radio heads (RRH). Specifically, we consider the uplink transmission where each RRH quantizes the received signals from the users by either point-to-point compression or Wyner-Ziv compression and then transmits the quantization bits to the BBU pool through capacity limited fronthhual links. To maximize the uplink sum rate, we jointly optimize the passive beamformers of IRSs and the quantization noise covariance matrices of fronthoul compression. An joint fronthaul compression and passive beamforming design is proposed by exploiting the Arimoto-Blahut algorithm and semidefinte relaxation (SDR). Numerical results show the performance gain achieved by the proposed algorithm.

eess.SP↗
arXiv2021

Exploring Targeted Universal Adversarial Perturbations to End-to-end ASR Models

Although end-to-end automatic speech recognition (e2e ASR) models are widely deployed in many applications, there have been very few studies to understand models' robustness against adversarial perturbations. In this paper, we explore whether a targeted universal perturbation vector exists for e2e ASR models. Our goal is to find perturbations that can mislead the models to predict the given targeted transcript such as "thank you" or empty string on any input utterance. We study two different attacks, namely additive and prepending perturbations, and their performances on the state-of-the-art LAS, CTC and RNN-T models. We find that LAS is the most vulnerable to perturbations among the three models. RNN-T is more robust against additive perturbations, especially on long utterances. And CTC is robust against both additive and prepending perturbations. To attack RNN-T, we find prepending perturbation is more effective than the additive perturbation, and can mislead the models to predict the same short target on utterances of arbitrary length.

eess.AS↗
arXiv2021

Electronic Properties of Single-Layer CoO$_2$/Au(111)

We report direct measurements via angle-resolved photoemission spectroscopy (ARPES) of the electronic dispersion of single-layer CoO$_2$. The Fermi contour consists of a large hole pocket centered at the $\overlineΓ$ point. To interpret the ARPES results, we use density functional theory (DFT) in combination with the multi-orbital Gutzwiller Approximation (DFT+GA), basing our calculations on crystalline structure parameters derived from x-ray photoelectron diffraction and low-energy electron diffraction. Our calculations are in good agreement with the measured dispersion. We conclude that the material is a moderately correlated metal. We also discuss substrate effects, and the influence of hydroxylation on the CoO$_2$ single-layer electronic structure.

cond-mat.mtrl-sci↗
arXiv2021

Deciphering Star Cluster Evolution by Shape Morphology

We analyze the morphological evolution of open clusters and provide shape parameters for 265 open clusters. The results show that the overall shape of sample clusters becomes more elliptical as they grow older, while their core remains circular or slightly trend to circularize. There is a negative correlation of the ellipticities with the number of members of the sample clusters. A significant negative correlation between the overall ellipticities and masses is also detected for the sample clusters with log(age/year) $\geq$ 8, suggesting that the overall shapes of the clusters are possibly influenced by the number of members and masses, in addition to the external forces and the surrounding environment. For most young sample clusters, the radial stratification degree of the short axis direction is greater than that of the long, implying that the radial stratification degree in the two directions within the young sample cluster may be unevenly affected by an internal evolutionary process. Older sample clusters exhibit lower stratification in the tangential direction, which possibly means those clusters may continue to survive for a long time at a low level of stratification. Our analysis shows that the overall shape of the sample clusters may be more susceptible to the influence of Galactic tides toward the Galactic center than the shear forces embedded in Galactic differential rotation. By analyzing the distribution of the ages and number of members of star clusters, we suggest that NGC 6791 may originate from superclusters.

astro-ph.GA↗
arXiv2021

A Survey on Multi-Task Learning

Multi-Task Learning (MTL) is a learning paradigm in machine learning and its aim is to leverage useful information contained in multiple related tasks to help improve the generalization performance of all the tasks. In this paper, we give a survey for MTL from the perspective of algorithmic modeling, applications and theoretical analyses. For algorithmic modeling, we give a definition of MTL and then classify different MTL algorithms into five categories, including feature learning approach, low-rank approach, task clustering approach, task relation learning approach and decomposition approach as well as discussing the characteristics of each approach. In order to improve the performance of learning tasks further, MTL can be combined with other learning paradigms including semi-supervised learning, active learning, unsupervised learning, reinforcement learning, multi-view learning and graphical models. When the number of tasks is large or the data dimensionality is high, we review online, parallel and distributed MTL models as well as dimensionality reduction and feature hashing to reveal their computational and storage advantages. Many real-world applications use MTL to boost their performance and we review representative works in this paper. Finally, we present theoretical analyses and discuss several future directions for MTL.

cs.LG↗
arXiv2021

On the limits of algorithmic prediction across the globe

The impact of predictive algorithms on people's lives and livelihoods has been noted in medicine, criminal justice, finance, hiring and admissions. Most of these algorithms are developed using data and human capital from highly developed nations. We tested how well predictive models of human behavior trained in a developed country generalize to people in less developed countries by modeling global variation in 200 predictors of academic achievement on nationally representative student data for 65 countries. Here we show that state-of-the-art machine learning models trained on data from the United States can predict achievement with high accuracy and generalize to other developed countries with comparable accuracy. However, accuracy drops linearly with national development due to global variation in the importance of different achievement predictors, providing a useful heuristic for policymakers. Training the same model on national data yields high accuracy in every country, which highlights the value of local data collection.

cs.LG↗
arXiv2021

Electron-electron interactions and weak anti-localization in few-layer ZrTe5 devices

Much effort has been devoted to the electronic properties of relatively thick ZrTe5 crystals, focusing on their three-dimensional topological effects. Thin ZrTe5 crystals, on the other hand, were much less explored experimentally. Here we present detailed magnetotransport studies of few-layer ZrTe5 devices, in which electron-electron interactions and weak anti-localization are observed. The coexistence of the two effects manifests themselves in corroborating evidence presented in the temperature and magnetic field dependence of the resistance. Notably, the temperature-dependent phase coherence length extracted from weak anti-localization agrees with strong electron-electron scattering in the sample. Meanwhile, universal conductance fluctuations have temperature and gate voltage dependence that is similar to that of the phase coherence length. Lastly, all the transport properties in thin ZrTe5 crystals show strong two-dimensional characteristics. Our results provide new insight into the highly intricate properties of topological material ZrTe5.

cond-mat.mtrl-sci↗
arXiv2021

Joint Transmit Precoding and Reflect Beamforming Design for IRS-Assisted MIMO Cognitive Radio Systems

Cognitive radio (CR) is an effective solution to improve the spectral efficiency (SE) of wireless communications by allowing the secondary users (SUs) to share spectrum with primary users (PUs). Meanwhile, intelligent reflecting surface (IRS) has been recently proposed as a promising approach to enhance SE and energy efficiency (EE) of wireless communication systems through intelligently reconfiguring the channel environment. In this paper, we consider an IRS-assisted downlink CR system, in which a secondary access point (SAP) communicates with multiple SUs without affecting multiple PUs in the primary network and all nodes are equipped with multiple antennas. Our design objective is to maximize the achievable weighted sum rate (WSR) of SUs subject to the total transmit power constraint at the SAP and the interference constraints at PUs, by jointly optimizing the transmit precoding at the SAP and the reflecting coefficients at the IRS. To deal with the complex objective function, the problem is reformulated by employing the well-known weighted minimum mean-square error (WMMSE) method and an alternating optimization (AO)-based algorithm is proposed. Furthermore, a special scenario with only one PU is considered and AO algorithm is adopted again. It is worth mentioning that the proposed algorithm has a much lower computational complexity than the above algorithm without the performance loss. Finally, some numerical simulations have been provided to demonstrate that the proposed algorithm outperforms other benchmark schemes.

cs.IT↗
arXiv2021

Learning Word-Level Confidence For Subword End-to-End ASR

We study the problem of word-level confidence estimation in subword-based end-to-end (E2E) models for automatic speech recognition (ASR). Although prior works have proposed training auxiliary confidence models for ASR systems, they do not extend naturally to systems that operate on word-pieces (WP) as their vocabulary. In particular, ground truth WP correctness labels are needed for training confidence models, but the non-unique tokenization from word to WP causes inaccurate labels to be generated. This paper proposes and studies two confidence models of increasing complexity to solve this problem. The final model uses self-attention to directly learn word-level confidence without needing subword tokenization, and exploits full context features from multiple hypotheses to improve confidence accuracy. Experiments on Voice Search and long-tail test sets show standard metrics (e.g., NCE, AUC, RMSE) improving substantially. The proposed confidence module also enables a model selection approach to combine an on-device E2E model with a hybrid model on the server to address the rare word recognition problem for the E2E model.

eess.AS↗
arXiv2021

Phase discontinuities induced scintillation enhancement: coherent vortex beams propagating through weak oceanic turbulence

Under the impact of an infinitely extended edge phase dislocation, optical vortices (screw phase dislocations) induce scintillation enhancement. The scintillation index of a beam consisting of two Gaussian vortex beams with ${\pm{1}}$ topological charges through weak oceanic turbulence is researched via derivation and phase screen simulation. Different combinations of two types of phase discontinuities can be obtained by changing the overlapping degree and the phase difference of two coherent Gaussian vortex beams. The scintillation indexes for them verify that the formation condition of the phenomenon is the coexistence of two types of phase discontinuities. And the enhanced scintillation index can be several orders of magnitude larger than that of a plane wave under weak perturbation (Rytov variance). This phenomenon could be useful for both optical vortex detection and perturbation measurement.

physics.optics↗
arXiv2021

Strange electrical transport: Colossal magnetoresistance via avoiding fully polarized magnetization in ferrimagnetic insulator Mn3Si2Te6

Colossal magnetoresistance is of great fundamental and technological significance and exists mostly in the manganites and a few other materials. Here we report colossal magnetoresistance that is starkly different from that in all other materials. The stoichiometric Mn3Si2Te6 is an insulator featuring a ferrimagnetic transition at 78 K. The resistivity drops by 7 orders of magnitude with an applied magnetic field above 9 Tesla, leading to an insulator-metal transition at up to 130 K. However, the colossal magnetoresistance occurs only when the magnetic field is applied along the magnetic hard axis and is surprisingly absent when the magnetic field is applied along the magnetic easy axis where magnetization is fully saturated. The anisotropy field separating the easy and hard axes is 13 Tesla, unexpected for the Mn ions with nominally negligible orbital momentum and spin-orbit interactions. Double exchange and Jahn-Teller distortions that drive the hole-doped manganites do not exist in Mn3Si2Te6. The phenomena fit no existing models, suggesting a unique, intriguing type of electrical transport.

cond-mat.str-el↗
arXiv2021

Self-supervised Low Light Image Enhancement and Denoising

This paper proposes a self-supervised low light image enhancement method based on deep learning, which can improve the image contrast and reduce noise at the same time to avoid the blur caused by pre-/post-denoising. The method contains two deep sub-networks, an Image Contrast Enhancement Network (ICE-Net) and a Re-Enhancement and Denoising Network (RED-Net). The ICE-Net takes the low light image as input and produces a contrast enhanced image. The RED-Net takes the result of ICE-Net and the low light image as input, and can re-enhance the low light image and denoise at the same time. Both of the networks can be trained with low light images only, which is achieved by a Maximum Entropy based Retinex (ME-Retinex) model and an assumption that noises are independently distributed. In the ME-Retinex model, a new constraint on the reflectance image is introduced that the maximum channel of the reflectance image conforms to the maximum channel of the low light image and its entropy should be the largest, which converts the decomposition of reflectance and illumination in Retinex model to a non-ill-conditioned problem and allows the ICE-Net to be trained with a self-supervised way. The loss functions of RED-Net are carefully formulated to separate the noises and details during training, and they are based on the idea that, if noises are independently distributed, after the processing of smoothing filters (\eg mean filter), the gradient of the noise part should be smaller than the gradient of the detail part. It can be proved qualitatively and quantitatively through experiments that the proposed method is efficient.

cs.CV↗
arXiv2021

Effects of $N(2000){5/2}^+$ on $γp \to K^+ Λ(1405)$

The photoproduction reaction of $γp \to K^+Λ(1405)$ is investigated based on an effective Lagrangian approach at the tree-level approximation with the purpose of understanding the reaction mechanism and extracting the resonance contents and the associated resonance parameters in this reaction. Apart from the $t$-channel $K$ and $K^\ast$ exchanges, $s$-channel nucleon ($N$) exchange, $u$-channel $Σ$, $Λ$, and $Λ(1405)$ exchanges, and generalized contact term, the exchanges of a minimum number of $N$ resonances in the $s$ channel are taken into account in constructing the reaction amplitudes to describe the experimental data. It is found that by introducing the $N(2000){5/2}^+$ resonance exchange in the $s$ channel, one can reproduce the most recent differential cross-section data from the CLAS Collaboration quite well. Further analysis shows that the cross sections of $γp \to K^+Λ(1405)$ at high energies are dominated by the $t$-channel $K$ exchange, while the contributions from the $s$-channel $N$ and $N(2000){5/2}^+$ exchanges are rather significant to the cross sections in the near-threshold energy region. Predictions for the beam and target asymmetries for $γp \to K^+Λ(1405)$ are given.

hep-ph↗
arXiv2021

Reinforcement Learning of Beam Codebooks in Millimeter Wave and Terahertz MIMO Systems

Millimeter wave (mmWave) and terahertz MIMO systems rely on pre-defined beamforming codebooks for both initial access and data transmission. Being pre-defined, however, these codebooks are commonly not optimized for specific environments, user distributions, and/or possible hardware impairments. This leads to large codebook sizes with high beam training overhead which increases the initial access/tracking latency and makes it hard for these systems to support highly mobile applications. To overcome these limitations, this paper develops a deep reinforcement learning framework that learns how to iteratively optimize the codebook beam patterns (shapes) relying only on the receive power measurements and without requiring any explicit channel knowledge. The developed model learns how to autonomously adapt the beam patterns to best match the surrounding environment, user distribution, hardware impairments, and array geometry. Further, this approach does not require any knowledge about the channel, array geometry, RF hardware, or user positions. To reduce the learning time, the proposed model designs a novel Wolpertinger-variant architecture that is capable of efficiently searching for an optimal policy in a large discrete action space, which is important for large antenna arrays with quantized phase shifters. This complex-valued neural network architecture design respects the practical RF hardware constraints such as the constant-modulus and quantized phase shifter constraints. Simulation results based on the publicly available DeepMIMO dataset confirm the ability of the developed framework to learn near-optimal beam patterns for both line-of-sight (LOS) and non-LOS scenarios and for arrays with hardware impairments without requiring any channel knowledge.

cs.IT↗