arXiv subjects

Zhiyuan Li

Publications and source records attributed to Zhiyuan Li.

At least 325 records · Page 18Linked to original sources

arXiv2020

Calibrated Intervention and Containment of the COVID-19 Pandemic

Within a short period of time, COVID-19 grew into a world-wide pandemic. Transmission by pre-symptomatic and asymptomatic viral carriers rendered intervention and containment of the disease extremely challenging. Based on reported infection case studies, we construct an epidemiological model that focuses on transmission around the symptom onset. The model is calibrated against incubation period and pairwise transmission statistics during the initial outbreaks of the pandemic outside Wuhan with minimal non-pharmaceutical interventions. Mathematical treatment of the model yields explicit expressions for the size of latent and pre-symptomatic subpopulations during the exponential growth phase, with the local epidemic growth rate as input. We then explore reduction of the basic reproduction number R_0 through specific disease control measures such as contact tracing, testing, social distancing, wearing masks and sheltering in place. When these measures are implemented in combination, their effects on R_0 multiply. We also compare our model behaviour to the first wave of the COVID-19 spreading in various affected regions and highlight generic and less generic features of the pandemic development.

q-bio.PE↗

arXiv2020

Realisation of adiabatic and diabatic CZ gates in superconducting qubits coupled with a tunable coupler

High fidelity two-qubit gates are fundamental for scaling up the superconducting number. We use two qubits coupled via a frequency-tunable coupler which can adjust the coupling strength, and demonstrate the CZ gate using two different schemes, adiabatic and diabatic methods. The Clifford based Randomized Benchmarking (RB) method is used to assess and optimize the CZ gate fidelity. The fidelity of adiabatic and diabatic CZ gates are 99.53(8)% and 98.72(2)%, respectively. We also analyze the errors induced by the decoherence. Comparing to 30 ns duration time of adiabatic CZ gate, the duration time of diabatic CZ gate is 19 ns, revealing lower incoherence error rate $r'_{\rm{incoherent, int}}$ = 0.0197(5) than $r_{\rm{incoherent, int}}$ = 0.0223(3).

quant-ph↗

arXiv2020

A Herschel mapping of [C ii], [O i] and [O iii] lines from the circumnuclear region of M31

The circumnuclear region of M31, consisting of multiphase interstellar medium, provides a close-up view of the interaction of the central supermassive black hole and surrounding materials. Far-infrared (FIR) line structure lines and their flux ratios can be used as diagnostics of physical properties of the neutral gas in this region. Here we present the first FIR spectroscopic mapping of the circumnuclear region of M31 in [C ii] 158 um, [O i] 63 um and [O iii] 88 um lines with the Herschel Space Observatory, covering a ~500 x 500 pc (2' x 2') field. Significant emissions of all three lines are detected along the so-called nuclear spiral across the central kpc of M31. The velocity field under a spatial resolution of ~50 pc of the three lines are in broad consistency and also consistent with previous CO(3-2) line observations in the central region. Combined with existing [C ii] and CO(3-2) observations of five other fields targeting on the disk, we derived the radial distribution of [C ii]/CO(3-2) flux ratio, and found that this ratio is higher in the center than the disk, indicating a low gas density and strong radiation field in the central region. We also found that the [C ii]/FIR ratio in the central region is 5.4 (+-0.8) x 10^-3, which exhibits an increasing trend with the galactocentric radius, suggesting an increasing contribution from old stellar population to dust heating towards the center.

astro-ph.GA↗

arXiv2020

Characterize noise correlation and enhance coherence via qubit motion

The identification of spacial noise correlation is of critical importance in developing error-corrected quantum devices, but it has barely been studied so far. In this work, we utilize an effective new method called qubit motion, to efficiently determine the noise correlations between any pair of qubits in a 7-qubit superconducting quantum system. The noise correlations between the same pairs of qubits are also investigated when the qubits are at distinct operating frequencies. What's more, in this multi-qubit system with the presence of noise correlations, we demonstrate the enhancing effect of qubit motion on the coherence of logic qubit, and we propose a Motion-CPMG operation sequence to more efficiently protect the logic state from decoherence, which is experimentally demonstrated to extend the decoherence time of logic qubit by nearly one order of magnitude.

quant-ph↗

arXiv2020

Moduli space of quasi-polarized K3 surfaces of degree 6 and 8

In this paper, we study the moduli space of quasi-polarized complex K3 surfaces of degree 6 and 8 via geometric invariant theory. The general members in such moduli spaces are complete intersections in projective spaces and we have natural GIT constructions for the corresponding moduli spaces and we show that the K3 surfaces with at worst ADE singularities are GIT stable. We give a concrete description of boundary of the compactification of the degree 6 case via the Hilbert-Mumford criterion. We compute the Picard group via Noether-Lefschetz theory and discuss the connection to the Looijenga's compactifications from arithmetic perspective. One of the main ingredients is the study of the projective models of K3 surfaces in terms of Noether-Lefschetz divisors.

math.AG↗

arXiv2020

Reconciling Modern Deep Learning with Traditional Optimization Analyses: The Intrinsic Learning Rate

Recent works (e.g., (Li and Arora, 2020)) suggest that the use of popular normalization schemes (including Batch Normalization) in today's deep learning can move it far from a traditional optimization viewpoint, e.g., use of exponentially increasing learning rates. The current paper highlights other ways in which behavior of normalized nets departs from traditional viewpoints, and then initiates a formal framework for studying their mathematics via suitable adaptation of the conventional framework namely, modeling SGD-induced training trajectory via a suitable stochastic differential equation (SDE) with a noise term that captures gradient noise. This yields: (a) A new ' intrinsic learning rate' parameter that is the product of the normal learning rate and weight decay factor. Analysis of the SDE shows how the effective speed of learning varies and equilibrates over time under the control of intrinsic LR. (b) A challenge -- via theory and experiments -- to popular belief that good generalization requires large learning rates at the start of training. (c) New experiments, backed by mathematical intuition, suggesting the number of steps to equilibrium (in function space) scales as the inverse of the intrinsic learning rate, as opposed to the exponential time convergence bound implied by SDE analysis. We name it the Fast Equilibrium Conjecture and suggest it holds the key to why Batch Normalization is effective.

cs.LG↗

arXiv2020

Exploring the Mass Segregation Effect of X-ray Sources in Globular Clusters. IV. Evidence of Black Hole Burning in $ω$ Centauri

Using X-ray sources as sensitive probes of stellar dynamical interactions in globular clusters (GCs), we study the mass segregation and binary burning processes in $ω$ Cen. We show that the mass segregation of X-ray sources is quenched in $ω$ Cen, while the X-ray source abundance of $ω$ Cen is much smaller than other GCs, and the binary hardness ratio (defined as $L_{\rm X}/(L_{\rm K}f_{b})$, with $f_{b}$ the binary fraction, $L_{\rm X}$ and $L_{\rm K}$ the cumulative X-ray and K band luminosity of GCs, respectively) of $ω$ Cen is located far below the $L_{\rm X}/(L_{\rm K}f_{b})-σ_{c}$ correlation line of the dynamically normal GCs. These evidences suggest that the binary burning processes are highly suppressed in $ω$ Cen, and other "heating mechanisms", very likely a black hole subsystem (BHS), are essential in the dynamical evolution of $ω$ Cen. Through the "black hole burning" processes (i.e., dynamical hardening of the BH binaries), the BHS is dominating the energy production of $ω$ Cen, which also makes $ω$ Cen a promising factory of gravitational-wave sources in the Galaxy.

astro-ph.HE↗

arXiv2020

Simple and Effective Regularization Methods for Training on Noisily Labeled Data with Generalization Guarantee

Over-parameterized deep neural networks trained by simple first-order methods are known to be able to fit any labeling of data. Such over-fitting ability hinders generalization when mislabeled training examples are present. On the other hand, simple regularization methods like early-stopping can often achieve highly nontrivial performance on clean test data in these scenarios, a phenomenon not theoretically understood. This paper proposes and analyzes two simple and intuitive regularization methods: (i) regularization by the distance between the network parameters to initialization, and (ii) adding a trainable auxiliary variable to the network output for each training example. Theoretically, we prove that gradient descent training with either of these two methods leads to a generalization guarantee on the clean data distribution despite being trained using noisy labels. Our generalization analysis relies on the connection between wide neural network and neural tangent kernel (NTK). The generalization bound is independent of the network size, and is comparable to the bound one can get when there is no label noise. Experimental results verify the effectiveness of these methods on noisily labeled datasets.

cs.LG↗

arXiv2020

Gravitational Test Beyond the First Post-Newtonian Order with the Shadow of the M87 Black Hole

The 2017 Event Horizon Telescope (EHT) observations of the central source in M87 have led to the first measurement of the size of a black-hole shadow. This observation offers a new and clean gravitational test of the black-hole metric in the strong-field regime. We show analytically that spacetimes that deviate from the Kerr metric but satisfy weak-field tests can lead to large deviations in the predicted black-hole shadows that are inconsistent with even the current EHT measurements. We use numerical calculations of regular, parametric, non-Kerr metrics to identify the common characteristic among these different parametrizations that control the predicted shadow size. We show that the shadow-size measurements place significant constraints on deviation parameters that control the second post-Newtonian and higher orders of each metric and are, therefore, inaccessible to weak-field tests. The new constraints are complementary to those imposed by observations of gravitational waves from stellar-mass sources.

gr-qc↗

arXiv2020

Periodic X-ray Sources in the Galactic Bulge: Application of the Gregory-Loredo Algorithm

We present a systematic study of periodic X-ray sources in the Limiting Window (LW), a $\sim$70 arcmin$^2$ field representative of the inner Galactic bulge and the target of $\sim$1 Ms Chandra observations. Using the Gregory-Loredo algorithm, which applies Bayes's theorem to the phase-folded light curve and is well-suited for irregularly sampled X-ray data, we detect 25 periodic signals in 23 discrete sources, among which 15 signals are new discoveries and two sources show dual periods. The vast majority of the 23 periodic sources are classified as magnetic cataclysmic variables (CVs), based on their period range, X-ray luminosities, spectral properties, and phase-folded light curves that are characteristic of spin modulation. Meanwhile, there is a paucity of non-magnetic CVs seen as periodic sources, which can be understood as due to a low detection efficiency for eclipsing sources. Under reasonable assumptions about the geometry of magnetic CVs and a large set of simulated X-ray light curves, we estimate the fraction of magnetic CVs in the inner Galactic bulge to be $\lesssim$23%, which is similar to that in the solar neighborhood. There is an apparent lack of long-period ($\gtrsim$3.3 hours) CVs in the LW, when contrasted with the range of known CVs in the solar neighborhood. We suggest that this might be an age effect, in the sense that CVs in the inner bulge are more evolved systems and have substantially shrunk their orbits.

astro-ph.HE↗

arXiv2020

Learning Geometry-Dependent and Physics-Based Inverse Image Reconstruction

Deep neural networks have shown great potential in image reconstruction problems in Euclidean space. However, many reconstruction problems involve imaging physics that are dependent on the underlying non-Euclidean geometry. In this paper, we present a new approach to learn inverse imaging that exploit the underlying geometry and physics. We first introduce a non-Euclidean encoding-decoding network that allows us to describe the unknown and measurement variables over their respective geometrical domains. We then learn the geometry-dependent physics in between the two domains by explicitly modeling it via a bipartite graph over the graphical embedding of the two geometry. We applied the presented network to reconstructing electrical activity on the heart surface from body-surface potential. In a series of generalization tasks with increasing difficulty, we demonstrated the improved ability of the presented network to generalize across geometrical changes underlying the data in comparison to its Euclidean alternatives.

eess.IV↗

arXiv2020

A Chandra X-ray Survey of Optically Selected AGN Pairs

We present a Chandra archival study of optically selected AGN pairs at a median redshift $\bar{z} \sim$ 0.1. Out of 1286 AGN pairs (with projected separations $r_p < 100$ kpc and velocity offsets $Δv < 600$ km s$^{-1}$) optically identified from the Sloan Digital Sky Survey Seventh Data Release, we find 67 systems with archival Chandra observations, which represent the largest sample of optically selected AGN pairs studied in the X-ray. Among the 67 AGN pairs, 21 systems have both nuclei detected in the X-rays, 36 have one nucleus detected in the X-rays, whereas 10 have no X-ray detection. The X-ray detection rate, 78/134 = 58% ($\pm 7$% 1$σ$ Poisson errors), is significantly higher than that (23/134 = $17\% \pm 4\%$) of a comparison sample of star-forming galaxy pairs, lending support to the optical AGN classification. In the conservative case where X-ray contamination from star formation is removed, the X-ray detection rate becomes $27\% \pm 4\%$, consistent with predictions from latest galaxy merger simulations. The 2-10 keV X-ray luminosity $L_{2-10~{\rm keV}}$ increases with decreasing projected separation in AGN pairs for $r_p \gtrsim 15$ kpc, suggesting an enhancement of black hole accretion even in early-stage mergers. On the other hand, $L_{2-10~{\rm keV}}$ appears to decrease with decreasing projected separation at $r_p \lesssim 15$ kpc, which is contradictory to predictions from merger simulations. The apparent decrease in $L_{2-10~{\rm keV}}$ of AGN pairs at $r_p \lesssim 15$ kpc may be caused by: (i) enhanced absorbing columns from merger-induced gas inflows, (ii) feedback effects from early-stage mergers, and/or (iii) small number statistics. Future X-ray studies with larger samples are needed to put our results on firmer statistical ground.

astro-ph.GA↗

arXiv2020

Well-posedness for the backward problems in time for general time-fractional diffusion equation

In this article, we consider a partial differential equation with Caputo time-derivative: $\partial_t^αu + Au = F$ where $0< α< 1$ and $u$ satisfies the zero Dirichlet boundary condition. For a non-symmetric elliptic operator $-A$ of the second order and given $F$, we prove the well-posedness for the backward problem in time and our result generalizes the existing results assuming that $A$ is symmetric. The key is the perturbation argument and the completeness of the generalized eigenfunctions of the elliptic operator $A$.

math.AP↗

arXiv2020

Galactic Center IRS13E: Colliding Stellar Winds or an Intermediate Mass Black Hole?

A small cluster of massive stars residing in the Galactic center, collectively known as IRS13E, is of special interest due to its close proximity to Sgr A* and the possibility that an embedded intermediate-mass black hole (IMBH) binds its member stars. It has been suggested that colliding winds from two member stars, both classified as Wolf-Rayet type, are responsible for the observed X-ray, infrared and radio emission from IRS13E. We have conducted an in-depth study of the X-ray spatial, temporal and spectral properties of IRS13E, based on 5.6 Ms of ultra-deep Chandra observations obtained over 20 years. These X-ray observations show no significant evidence for source variability. We have also explored the kinematics of the cluster members, using Keck near-infrared imaging and spectroscopic data on a 14-yr baseline that considerably improve the accuracy of stars' proper motions. The observations are interpreted using 3-dimensional hydrodynamical simulations of colliding winds tailored to match the physical conditions of IRS13E, leading us to conclude that the observed X-ray spectrum and morphology can be well explained by the colliding wind scenario, in the meantime offering no support for the presence of a putative IMBH. An IMBH more massive than a few $10^3{\rm~M_\odot}$ is also strongly disfavored by the stellar kinematics.

astro-ph.HE↗

arXiv2020

Semi-supervised Medical Image Classification with Global Latent Mixing

Computer-aided diagnosis via deep learning relies on large-scale annotated data sets, which can be costly when involving expert knowledge. Semi-supervised learning (SSL) mitigates this challenge by leveraging unlabeled data. One effective SSL approach is to regularize the local smoothness of neural functions via perturbations around single data points. In this work, we argue that regularizing the global smoothness of neural functions by filling the void in between data points can further improve SSL. We present a novel SSL approach that trains the neural network on linear mixing of labeled and unlabeled data, at both the input and latent space in order to regularize different portions of the network. We evaluated the presented model on two distinct medical image data sets for semi-supervised classification of thoracic disease and skin lesion, demonstrating its improved performance over SSL with local perturbations and SSL with global mixing but at the input space only. Our code is available at https://github.com/Prasanna1991/LatentMixing.

cs.LG↗

arXiv2020

What has quenched the massive spiral galaxies?

Quenched massive spiral galaxies have attracted great attention recently, as more data is available to constrain their environment and cold gas content. However, the quenching mechanism is still uncertain, as it depends on the mass range and baryon budget of the galaxy. In this letter, we report the identification of a rare population of very massive, quenched spiral galaxies with stellar mass $\gtrsim10^{11}{\rm~M_\odot}$ and halo mass $\gtrsim10^{13}{\rm~M_\odot}$ from the Sloan Digital Sky Survey at redshift $z\sim0.1$. Our CO observations using the IRAM-30m telescope show that these galaxies contain only a small amount of molecular gas. Similar galaxies are also seen in the state-of-the-art semi-analytical models and hydro-dynamical simulations. It is found from these theoretical models that these quenched spiral galaxies harbor massive black holes, suggesting that feedback from the central black holes has quenched these spiral galaxies. This quenching mechanism seems to challenge the popular scenario of the co-evolution between massive black holes and massive bulges.

astro-ph.GA↗

arXiv2020

Resolving the Nuclear Radio Emission from M32 with Very Large Array

The Local Group dwarf elliptical galaxy M32 hosts one of the nearest and most under-luminous super-massive black holes (SMBHs) ever known, offering a rare opportunity to study the physics of accreting SMBHs at the most quiescent state. Recent Very Large Array (VLA) observations have detected a radio source at the nucleus of M32, which is suggested to be the radio counterpart of the SMBH. To further investigate the radio properties of this nuclear source, we have conducted follow-up, high-resolution VLA observations in four epochs between 2015--2017, each with dual frequencies. At 6 GHz, the nuclear source is resolved under an angular resolution of $\sim$0\farcs4, exhibiting a coreless, slightly lopsided morphology with a detectable extent of $\sim$2.5 \arcsec ($\sim$10 parsec). No significant variability can be found among the four epochs. At 15 GHz, no significant emission can be detected within the same region, pointing to a steep intrinsic radio spectrum (with a 3\,$σ$ upper limit of -1.46 for the spectral index). We discuss possible scenarios for the nature of this nuclear source and conclude that a stellar origin, in particular planetary nebulae, X-ray binaries, supernova remnants or diffuse ionized gas powered by massive stars, can be ruled out.Instead, the observed radio properties can be explained by synchrotron radiation from a hypothetical wind driven by the weakly accreting SMBH.

astro-ph.GA↗

arXiv2020

Unique determination of fractional order and source term in a fractional diffusion equation from sparse boundary data

In this article, for a two dimensional fractional diffusion equation, we study an inverse problem for simultaneous restoration of the fractional order and the source term from the sparse boundary measurements. By the adjoint system corresponding to our diffusion equation, we construct useful quantitative relation between unknowns and measurements. From Laplace transform and the knowledge in complex analysis, the uniqueness theorem is proved.

math.AP↗