arXiv subjects

Jie Yu

Publications and source records attributed to Jie Yu.

At least 109 records · Page 6Linked to original sources

arXiv2022

ET White Paper: To Find the First Earth 2.0

We propose to develop a wide-field and ultra-high-precision photometric survey mission, temporarily named "Earth 2.0 (ET)". This mission is designed to measure, for the first time, the occurrence rate and the orbital distributions of Earth-sized planets. ET consists of seven 30cm telescopes, to be launched to the Earth-Sun's L2 point. Six of these are transit telescopes with a field of view of 500 square degrees. Staring in the direction that encompasses the original Kepler field for four continuous years, this monitoring will return tens of thousands of transiting planets, including the elusive Earth twins orbiting solar-type stars. The seventh telescope is a 30cm microlensing telescope that will monitor an area of 4 square degrees toward the galactic bulge. This, combined with simultaneous ground-based KMTNet observations, will measure masses for hundreds of long-period and free-floating planets. Together, the transit and the microlensing telescopes will revolutionize our understandings of terrestrial planets across a large swath of orbital distances and free space. In addition, the survey data will also facilitate studies in the fields of asteroseismology, Galactic archeology, time-domain sciences, and black holes in binaries.

astro-ph.IM↗

arXiv2022

Incorporation of density scaling constraint in density functional design via contrastive representation learning

In a data-driven paradigm, machine learning (ML) is the central component for developing accurate and universal exchange-correlation (XC) functionals in density functional theory (DFT). It is well known that XC functionals must satisfy several exact conditions and physical constraints, such as density scaling, spin scaling, and derivative discontinuity. In this work, we demonstrate that contrastive learning is a computationally efficient and flexible method to incorporate a physical constraint in ML-based density functional design. We propose a schematic approach to incorporate the uniform density scaling property of electron density for exchange energies by adopting contrastive representation learning during the pretraining task. The pretrained hidden representation is transferred to the downstream task to predict the exchange energies calculated by DFT. The electron density encoder transferred from the pretraining task based on contrastive learning predicts exchange energies that satisfy the scaling property, while the model trained without using contrastive learning gives poor predictions for the scaling-transformed electron density systems. Furthermore, the model with pretrained encoder gives a satisfactory performance with only small fractions of the whole augmented dataset labeled, comparable to the model trained from scratch using the whole dataset. The results demonstrate that incorporating exact constraints through contrastive learning can enhance the understanding of density-energy mapping using neural network (NN) models with less data labeling, which will be beneficial to generalizing the application of NN-based XC functionals in a wide range of scenarios that are not always available experimentally but theoretically justified. This work represents a viable pathway toward the machine learning design of a universal density functional via representation learning.

physics.chem-ph↗

arXiv2022

Wall stabilization of the rigid ballooning $m=1$ mode in a long-thin mirror trap

The prospect of stabilization of the $m=1$ ``rigid'' ballooning mode in an open axially symmetric long-thin trap with the help of a conducting lateral wall surrounding a column of isotropic plasma is studied. It is found that for effective wall stabilization, the beta parameter must exceed $70\%$. The dependence of the critical beta on the mirror ratio, the radial pressure profile, and the axial profile of the vacuum magnet has been studied. It is shown that when a conductive lateral wall is combined with conductive end plates simulating attachment of the end MHD stabilizers to the central cell of an open trap, there are two critical beta values and two stability zones that can merge, making stable the entire range of allowable beta values $0<β<1$.

physics.plasm-ph↗

arXiv2022

Approximating Nash Equilibrium for Production Control with Sticky Price

We study a mean field game problem arising from the production control for multiple firms with price stickiness in the commodity market. The price dynamics for each firm is described as a (controlled) jump-diffusion process with mean-field interaction. Each firm aims to maximize her expectation of cumulative net profit coupled with each other through price processes. By solving the limiting control problem and a fixed-point problem, we construct an explicit approximating Nash equilibrium when the number of firms grows large.

math.OC↗

arXiv2022

Ages of Main-Sequence Turn-Off Stars from the GALAH Survey

Main sequence turn-off (MSTO) stars are good tracers of Galactic populations since their ages can be reliably estimated from atmospheric parameters. Based on the GALAH survey, we use the Yale Rotation Evolution Code to determine ages of 2926 MSTO stars with mean age uncertainty $\sim$10\% considering the variation of C and O abundances. Ages of CO-poor stars are systematically affected by $\sim$10\% due to the C and O abundances, globally shifting to $\sim$0.5 Gyr older compared to the results using solar metal-mixture. Of the stars with \mbox{[Fe/H] $\sim$0.3-0.5} or \mbox{[O/Fe]~$\leq$~-0.25}, many have fractional age differences~$\geq$~20\%, and even reach up to 36\%. The age-metallicity relation appears to possibly exist two distinct sequences: a young sequence of stars with age mostly $<$ 7 Gyr, and a relatively older sequence of stars with age mostly $>$ 7 Gyr, overlapping at 5 Gyr $\leq$~age~$\leq$ 7 Gyr. Moreover, the trends of abundances to age ratios show two corresponding sequences, especially in [O/Fe]-age plane. We also find that [Y/Mg] is a good chemical clock in disk populations. The young sequence and the old sequence can not be separated based on chemistry or kinematics, therefore stellar age is an important parameter to distinguish these two sequences in our sample.

astro-ph.SR↗

arXiv2022

Modeling Stellar Oscillations and Granulation in Radial Velocity Time Series: A Fourier-based Method

Tens of thousands of solar-like oscillating stars have been observed by space missions. Their photometric variability in the Fourier domain can be parameterized by a sum of two super-Lorentizian functions for granulation and a Gaussian-shaped power excess for oscillation. The photometric granulation/oscillation parameters scale with stellar parameters and they can also make predictions for corresponding parameters in radial velocity measurements. Based on scaling relations, we simulate realistic radial velocity time series and examine how the root-mean-square scatter of radial velocity measurements varies with stellar parameters and different observation strategies such as the length of integration time and gaps in the time series. Using stars with extensive spectroscopic observations from the spectrographs (SONG and HARPS), we measure the granulation amplitude and timescale from the power spectrum of the radial velocity time series. We compare these measurements with literature values based on Kepler photometry. We find that the granulation amplitude in radial velocity can be well predicted from the photometry and scaling relations. Both granulation timescales in radial velocity agree with those predicted from photometry for giants and sub-giants. However, for main-sequence stars, only one granulation timescale in radial velocity is in agreement with the photometric-based values, while the other timescale generally lies at lower frequencies compared to the result of photometry. In conclusion, we show the photometric scaling relations from Kepler photometry and the scaling relationship to Doppler observations can be very useful for predicting the photometric and radial velocity stellar variabilities due to stellar granulation and oscillation.

astro-ph.SR↗

arXiv2021

Dynamic Graph Learning-Neural Network for Multivariate Time Series Modeling

Multivariate time series forecasting is a challenging task because the data involves a mixture of long- and short-term patterns, with dynamic spatio-temporal dependencies among variables. Existing graph neural networks (GNN) typically model multivariate relationships with a pre-defined spatial graph or learned fixed adjacency graph. It limits the application of GNN and fails to handle the above challenges. In this paper, we propose a novel framework, namely static- and dynamic-graph learning-neural network (SDGL). The model acquires static and dynamic graph matrices from data to model long- and short-term patterns respectively. Static matric is developed to capture the fixed long-term association pattern via node embeddings, and we leverage graph regularity for controlling the quality of the learned static graph. To capture dynamic dependencies among variables, we propose dynamic graphs learning method to generate time-varying matrices based on changing node features and static node embeddings. And in the method, we integrate the learned static graph information as inductive bias to construct dynamic graphs and local spatio-temporal patterns better. Extensive experiments are conducted on two traffic datasets with extra structural information and four time series datasets, which show that our approach achieves state-of-the-art performance on almost all datasets. If the paper is accepted, I will open the source code on github.

cs.LG↗

arXiv2021

Age Determination of Galaxy Merger Remnant Stars using Asteroseismology

The Milky Way was shaped by the mergers with several galaxies in the past. We search for remnant stars that were born in these foreign galaxies and assess their ages in an effort to put upper limits on the merger times and thereby better understand the evolutionary history of our Galaxy. Using 6D-phase space information from Gaia eDR3 and chemical information from APOGEE DR16, we kinematically and chemically select $23$ red giant stars belonging to former dwarf galaxies that merged with the Milky Way. With added asteroseismology from Kepler and K2, we determine the ages of the $23$ ex-situ stars and $55$ in-situ stars with great precision. We find that all the ex-situ stars are consistent with being older than $8$ Gyr. While it is not possible to associate all the stars with a specific dwarf galaxy we classify eight of them as Gaia-Enceladus/Sausage stars, which is one of the most massive mergers in our Galaxy's history. We determine their mean age to be $9.5^{+1.2}_{-1.3}$ Gyr consistent with a merger time of $8$-$10$ Gyr ago. The rest of the stars are possibly associated with Kraken, Thamnos, Sequoia, or another extragalactic progenitor. The age determination of ex-situ stars paves the way to more accurately pinning down when the merger events occurred and hence provide tight constraints useful for simulating how these events unfolded.

astro-ph.GA↗

arXiv2021

One Million Scenes for Autonomous Driving: ONCE Dataset

Current perception models in autonomous driving have become notorious for greatly relying on a mass of annotated data to cover unseen cases and address the long-tail problem. On the other hand, learning from unlabeled large-scale collected data and incrementally self-training powerful recognition models have received increasing attention and may become the solutions of next-generation industry-level powerful and robust perception models in autonomous driving. However, the research community generally suffered from data inadequacy of those essential real-world scene data, which hampers the future exploration of fully/semi/self-supervised methods for 3D perception. In this paper, we introduce the ONCE (One millioN sCenEs) dataset for 3D object detection in the autonomous driving scenario. The ONCE dataset consists of 1 million LiDAR scenes and 7 million corresponding camera images. The data is selected from 144 driving hours, which is 20x longer than the largest 3D autonomous driving dataset available (e.g. nuScenes and Waymo), and it is collected across a range of different areas, periods and weather conditions. To facilitate future research on exploiting unlabeled data for 3D detection, we additionally provide a benchmark in which we reproduce and evaluate a variety of self-supervised and semi-supervised methods on the ONCE dataset. We conduct extensive analyses on those methods and provide valuable observations on their performance related to the scale of used data. Data, code, and more information are available at https://once-for-auto-driving.github.io/index.html.

cs.CV↗

arXiv2021

A binary with a $δ$~Scuti star and an oscillating red giant: orbit and asteroseismology of KIC9773821

We study the $δ$ Scuti -- red giant binary KIC9773821, the first double-pulsator binary of its kind. It was observed by \textit{Kepler} during its four-year mission. Our aims are to ascertain whether the system is bound, rather than a chance alignment, and to identify the evolutionary state of the red giant via asteroseismology. An extension of these aims is to determine a dynamical mass and an age prior for a $δ$ Sct star, which may permit mode identification via further asteroseismic modelling. We determine spectroscopic parameters and radial velocities (RVs) for the red giant component using HERMES@Mercator spectroscopy. Light arrival-time delays from the $δ$ Sct pulsations are used with the red-giant RVs to determine that the system is bound and to infer its orbital parameters, including the binary mass ratio. We use asteroseismology to model the individual frequencies of the red giant to give a mass of $2.10^{+0.20}_{-0.10}$ M$_{\odot}$ and an age of $1.08^{+0.06}_{-0.24}$ Gyr. We find that it is a helium-burning secondary clump star, confirm that it follows the standard $ν_{\rm max}$ scaling relation, and confirm its observed period spacings match their theoretical counterparts in the modelling code MESA. Our results also constrain the mass and age of the $δ$ Sct star. We leverage these constraints to construct $δ$ Sct models in a reduced parameter space and identify four of its five pulsation modes.

astro-ph.SR↗

arXiv2021

Incorporating dependence on ice thickness in empirical parameterizations of wave dissipation by sea ice

This study is part of an effort to improve the Navy's ability to forecast wind-generated ocean waves in ice-infested regions, and here we are attempting to further this goal by improving prediction of dissipation of wave energy by sea ice. Rogers et al. (2021) presented new estimates of frequency-dependent dissipation of wave energy by sea ice, based on model-data inversion, and studied the correlation with various other parameters, such as ice thickness and sea state variables. Here, we use that dataset to propose a new dissipation parameterization which explicitly incorporates the dependence on the ice thickness, in addition to the wave frequency. The goal is to determine whether a parameterization dependent on wave frequency and ice thickness can be more accurate than one dependent only on wave frequency. Due to the dominant impact of frequency and confounding difficulties of field measurements, this is not a foregone conclusion. A parameterization is developed using the non-dimensionalization approach proposed by Yu et al. (2019). We find that the non-dimensionalization does result in significant scale collapse of the data, and inclusion of ice thickness does improve accuracy, most evidenced by reduced scatter when applied to the same dataset. However, evaluations against independent datasets are mixed. Possible reasons for this are discussed.

physics.ao-ph↗

arXiv2020

Asteroseismology of luminous red giants with Kepler. II. Dependence of mass loss on pulsations and radiation

Mass loss by red giants is an important process to understand the final stages of stellar evolution and the chemical enrichment of the interstellar medium. Mass-loss rates are thought to be controlled by pulsation-enhanced dust-driven outflows. Here we investigate the relationships between mass loss, pulsations, and radiation, using 3213 luminous Kepler red giants and 135000 ASAS-SN semiregulars and Miras. Mass-loss rates are traced by infrared colours using 2MASS and WISE and by observed-to-model WISE fluxes, and are also estimated using dust mass-loss rates from literature assuming a typical gas-to-dust mass ratio of 400. To specify the pulsations, we extract the period and height of the highest peak in the power spectrum of oscillation. Absolute magnitudes are obtained from the 2MASS Ks band and the Gaia DR2 parallaxes. Our results follow. (i) Substantial mass loss sets in at pulsation periods above ~60 and ~100 days, corresponding to Asymptotic-Giant-Branch stars at the base of the period-luminosity sequences C' and C. (ii) The mass-loss rate starts to rapidly increase in semiregulars for which the luminosity is just above the Red-Giant-Branch tip and gradually plateaus to a level similar to that of Miras. (iii) The mass-loss rates in Miras do not depend on luminosity, consistent with pulsation-enhanced dust-driven winds. (iv) The accumulated mass loss on the Red Giant Branch consistent with asteroseismic predictions reduces the masses of red-clump stars by 6.3%, less than the typical uncertainty on their asteroseismic masses. Thus mass loss is currently not a limitation of stellar age estimates for galactic archaeology studies.

astro-ph.SR↗

arXiv2020

Simple hydrogenic estimates for the exchange and correlation energies of atoms and atomic ions, with implications for density functional theory

Exact density functionals for the exchange and correlation energies are approximated in practical calculations for the ground-state electronic structure of a many-electron system. An important exact constraint for the construction of approximations is to recover the correct non-relativistic large-$Z$ expansions for the corresponding energies of neutral atoms with atomic number $Z$ and electron number $N=Z$, which are correct to leading order ($-0.221 Z^{5/3}$ and $-0.021 Z \ln Z$ respectively) even in the lowest-rung or local density approximation. We find that hydrogenic densities lead to $E_x(N,Z) \approx -0.354 N^{2/3} Z$ (as known before only for $Z \gg N \gg 1$) and $E_c \approx -0.02 N \ln N$. These asymptotic estimates are most correct for atomic ions with large $N$ and $Z \gg N$, but we find that they are qualitatively and semi-quantitatively correct even for small $N$ and for $N \approx Z$. The large-$N$ asymptotic behavior of the energy is pre-figured in small-$N$ atoms and atomic ions, supporting the argument that widely-predictive approximate density functionals should be designed to recover the correct asymptotics. It is shown that the exact Kohn-Sham correlation energy, when calculated from the pure ground-state wavefunction, should have no contribution proportional to $Z$ in the $Z\to \infty$ limit for any fixed $N$.

physics.chem-ph↗

arXiv2020

CardioLearn: A Cloud Deep Learning Service for Cardiac Disease Detection from Electrocardiogram

Electrocardiogram (ECG) is one of the most convenient and non-invasive tools for monitoring peoples' heart condition, which can use for diagnosing a wide range of heart diseases, including Cardiac Arrhythmia, Acute Coronary Syndrome, et al. However, traditional ECG disease detection models show substantial rates of misdiagnosis due to the limitations of the abilities of extracted features. Recent deep learning methods have shown significant advantages, but they do not provide publicly available services for those who have no training data or computational resources. In this paper, we demonstrate our work on building, training, and serving such out-of-the-box cloud deep learning service for cardiac disease detection from ECG named CardioLearn. The analytic ability of any other ECG recording devices can be enhanced by connecting to the Internet and invoke our open API. As a practical example, we also design a portable smart hardware device along with an interactive mobile program, which can collect ECG and detect potential cardiac diseases anytime and anywhere.

eess.SP↗

arXiv2020

Two-dimensional MX family of Dirac materials with tunable electronic and topological properties

We propose a novel class of two-dimensional (2D) Dirac materials in the MX family (M=Be, Mg, Zn and Cd, X = Cl, Br and I), which exhibit graphene-like band structures with linearly-dispersing Dirac-cone states over large energy scales (0.8~1.8 eV) and ultra-high Fermi velocities comparable to graphene. The electronic and topological properties are found to be highly tunable and amenable to effective modulation via anion-layer substitution and vertical electric field. The electronic structures of several members of the family are shown to host a Van-Hove singularity (VHS) close to the energy of the Dirac node. The enhanced density-of-states associated with these VHSs could provide a mechanism for inducing topological superconductivity. The presence of sizable band gaps, ultra-high carrier mobilities, and small effective masses makes the MX family promising for electronics and spintronics applications.

cond-mat.mtrl-sci↗

arXiv2020

Very regular high-frequency pulsation modes in young intermediate-mass stars

Asteroseismology is a powerful tool for probing the internal structures of stars by using their natural pulsation frequencies. It relies on identifying sequences of pulsation modes that can be compared with theoretical models, which has been done successfully for many classes of pulsators, including low-mass solar-type stars, red giants, high-mass stars and white dwarfs. However, a large group of pulsating stars of intermediate mass--the so-called delta Scuti stars--have rich pulsation spectra for which systematic mode identification has not hitherto been possible. This arises because only a seemingly random subset of possible modes are excited, and because rapid rotation tends to spoil the regular patterns. Here we report the detection of remarkably regular sequences of high-frequency pulsation modes in 60 intermediate-mass main-sequence stars, allowing definitive mode identification. Some of these stars have space motions that indicate they are members of known associations of young stars, and modelling of their pulsation spectra confirms that these stars are indeed young.

astro-ph.SR↗

arXiv2020

Solar-type Stars Observed by LAMOST and Kepler

Obtaining measurements of chromospheric and photometric activity of stars with near-solar fundamental parameters and rotation periods is important for a better understanding of solar-stellar connection. We select a sample of 2603 stars with near-solar fundamental parameters from the Large Sky Area Multi-Object Fiber Spectroscopic Telescope (LAMOST)-Kepler field and use LAMOST spectra to measure their chromospheric activity and Kepler light curves to measure their photospheric activity (i.e., the amplitude of the photometric variability). While the rotation periods of 1556 of these stars could not be measured due to the low amplitude of the photometric variability and highly irregular temporal profile of light curves, 254 stars were further identified as having near-solar rotation periods. We show that stars with near-solar rotation periods have chromospheric activities that are systematically higher than stars with undetected rotation periods. Furthermore, while the solar level of photospheric and chromospheric activity appears to be typical for stars with undetected rotation periods, the Sun appears to be less active than most stars with near-solar rotation periods (both in terms of photospheric and chromospheric activity).

astro-ph.SR↗

arXiv2020

Magnetic Activity of F-, G-, and K-type Stars in the LAMOST-Kepler Field

Monitoring chromospheric and photospheric indexes of magnetic activity can provide valuable information, especially the interaction between different parts of the atmosphere and their response to magnetic fields. We extract chromospheric indexes, S and Rhk+, for 59,816 stars from LAMOST spectra in the LAMOST-Kepler program, and photospheric index, Reff, for 5575 stars from Kepler light curves. The log Reff shows positive correlation with log Rhk+. We estimate the power-law indexes between Reff and Rhk+ for F-, G-, and K-type stars, respectively. We also confirm the dependence of both chromospheric and photospheric activity on stellar rotation. Ca II H and K emissions and photospheric variations generally decrease with increasing rotation periods for stars with rotation periods exceeding a few days. The power-law indexes in exponential decay regimes show different characteristics in the two activity-rotation relations. The updated largest sample including the activity proxies and reported rotation periods provides more information to understand the magnetic activity for cool stars.

astro-ph.SR↗