arXiv subjects

Yue Zhao

Publications and source records attributed to Yue Zhao.

At least 433 records · Page 24Linked to original sources

arXiv2019

Searching for Dark Photon Dark Matter in LIGO O1 Data

Dark matter exists in our Universe, but its nature remains mysterious. The remarkable sensitivity of the Laser Interferometer Gravitational-Wave Observatory (LIGO) may be able to solve this mystery. A good dark matter candidate is the ultralight dark photon. Because of its interaction with ordinary matter, it induces displacements on LIGO mirrors that can lead to an observable signal. In a study that bridges gravitational wave science and particle physics, we perform a direct dark matter search using data from LIGO's first (O1) data run, as opposed to an indirect search for dark matter via its production of gravitational waves. We demonstrate an achieved sensitivity on squared coupling as $\sim 4 \times 10^{-45}$, in a $U(1)_{\rm B}$ dark photon dark matter mass band around $m_{\rm A} \sim 4 \times 10^{-13}$eV. Substantially improved search sensitivity is expected during the coming years of continued data taking by LIGO and other gravitational wave detectors in a growing global network.

hep-ph↗

arXiv2019

Inverse scattering for the one-dimensional Helmholtz equation with piecewise constant wave speed

This paper analyzes inverse scattering for the one-dimensional Helmholtz equation in the case where the wave speed is piecewise constant. Scattering data recorded for an arbitrarily small interval of frequencies is shown to determine the wave speed uniquely, and a direct reconstruction algorithm is presented. The algorithm is exact provided data is recorded for a sufficiently wide range of frequencies and the jump points of the wave speed are equally spaced with respect to travel time. Numerical examples show that the algorithm works also in the general case of arbitrary wave speed (either with jumps or continuously varying etc.) giving progressively more accurate approximations as the range of recorded frequencies increases. A key underlying theoretical insight is to associate scattering data to compositions of automorphisms of the unit disk, which are in turn related to orthogonal polynomials on the unit circle. The algorithm exploits the three-term recurrence of orthogonal polynomials to reduce the required computation.

math.AP↗

arXiv2019

XGBOD: Improving Supervised Outlier Detection with Unsupervised Representation Learning

A new semi-supervised ensemble algorithm called XGBOD (Extreme Gradient Boosting Outlier Detection) is proposed, described and demonstrated for the enhanced detection of outliers from normal observations in various practical datasets. The proposed framework combines the strengths of both supervised and unsupervised machine learning methods by creating a hybrid approach that exploits each of their individual performance capabilities in outlier detection. XGBOD uses multiple unsupervised outlier mining algorithms to extract useful representations from the underlying data that augment the predictive capabilities of an embedded supervised classifier on an improved feature space. The novel approach is shown to provide superior performance in comparison to competing individual detectors, the full ensemble and two existing representation learning based algorithms across seven outlier datasets.

cs.LG↗

arXiv2019

On the Equilibria and Efficiency of Electricity Markets with Renewable Power Producers and Congestion Constraints

With increasing renewable penetration in power systems, a prominent challenge in efficient and reliable power system operation is handling the uncertainties inherent in the renewable generation. In this paper, we propose a simple two-settlement market mechanism in which renewable power producers (RPPs) participate, so that a) the independent system operator (ISO) does not need to consider the uncertainties of the renewables in its economic dispatch, and yet b) the market equilibrium is shown to approach social efficiency as if the ISO solves a stochastic optimization taking into account all the uncertainties. In showing this result, a key innovation is a new approach of efficiently computing the Nash equilibrium (NE) among the strategic RPPs in congestion-constrained power networks. In particular, the proposed approach decouples finding an NE into searching over congestion patterns and computing an NE candidate assuming a congestion pattern. As such, the computational complexity of finding an NE grows only cubically with the number of RPPs in the market. We demonstrate our results in the IEEE 14-bus system and show that the NE approaches social efficiency as the number of RPPs grows.

cs.GT↗

arXiv2019

DCSO: Dynamic Combination of Detector Scores for Outlier Ensembles

Selecting and combining the outlier scores of different base detectors used within outlier ensembles can be quite challenging in the absence of ground truth. In this paper, an unsupervised outlier detector combination framework called DCSO is proposed, demonstrated and assessed for the dynamic selection of most competent base detectors, with an emphasis on data locality. The proposed DCSO framework first defines the local region of a test instance by its k nearest neighbors and then identifies the top-performing base detectors within the local region. Experimental results on ten benchmark datasets demonstrate that DCSO provides consistent performance improvement over existing static combination approaches in mining outlying objects. To facilitate interpretability and reliability of the proposed method, DCSO is analyzed using both theoretical frameworks and visualization techniques, and presented alongside empirical parameter setting instructions that can be used to improve the overall performance.

cs.LG↗

arXiv2019

Combining Machine Learning Models using combo Library

Model combination, often regarded as a key sub-field of ensemble learning, has been widely used in both academic research and industry applications. To facilitate this process, we propose and implement an easy-to-use Python toolkit, combo, to aggregate models and scores under various scenarios, including classification, clustering, and anomaly detection. In a nutshell, combo provides a unified and consistent way to combine both raw and pretrained models from popular machine learning libraries, e.g., scikit-learn, XGBoost, and LightGBM. With accessibility and robustness in mind, combo is designed with detailed documentation, interactive examples, continuous integration, code coverage, and maintainability check; it can be installed easily through Python Package Index (PyPI) or https://github.com/yzhao062/combo.

cs.LG↗

arXiv2019

SynC: A Unified Framework for Generating Synthetic Population with Gaussian Copula

Synthetic population generation is the process of combining multiple socioeconomic and demographic datasets from different sources and/or granularity levels, and downscaling them to an individual level. Although it is a fundamental step for many data science tasks, an efficient and standard framework is absent. In this study, we propose a multi-stage framework called SynC (Synthetic Population via Gaussian Copula) to fill the gap. SynC first removes potential outliers in the data and then fits the filtered data with a Gaussian copula model to correctly capture dependencies and marginal distributions of sampled survey data. Finally, SynC leverages predictive models to merge datasets into one and then scales them accordingly to match the marginal constraints. We make three key contributions in this work: 1) propose a novel framework for generating individual level data from aggregated data sources by combining state-of-the-art machine learning and statistical techniques, 2) demonstrate its value as a feature engineering tool, as well as an alternative to data collection in situations where gathering is difficult through two real-world datasets, 3) release an easy-to-use framework implementation for reproducibility, and 4) ensure the methodology is scalable at the production level and can easily incorporate new data.

cs.LG↗

arXiv2019

Magnetic order induced polarization anomaly of Raman scattering in 2D magnet CrI$_3$

The recent discovery of 2D magnets has revealed various intriguing phenomena due to the coupling between spin and other degree of freedoms (such as helical photoluminescence, nonreciprocal SHG). Previous research on the spin-phonon coupling effect mainly focuses on the renormalization of phonon frequency. Here we demonstrate that the Raman polarization selection rules of optical phonons can be greatly modified by the magnetic ordering in 2D magnet CrI$_3$. For monolayer samples, the dominant A$\rm_{1g}$ peak shows abnormally high intensity in the cross polarization channel at low temperature, which is forbidden by the selection rule based on the lattice symmetry. While for bilayer, this peak is absent in the cross polarization channel for the layered antiferromagnetic (AFM) state and reappears when it is tuned to the ferromagnetic (FM) state by an external magnetic field. Our findings shed light on exploring the emergent magneto-optical effects in 2D magnets.

cond-mat.mes-hall↗

arXiv2019

Seeing isn't Believing: Practical Adversarial Attack Against Object Detectors

In this paper, we presented systematic solutions to build robust and practical AEs against real world object detectors. Particularly, for Hiding Attack (HA), we proposed the feature-interference reinforcement (FIR) method and the enhanced realistic constraints generation (ERG) to enhance robustness, and for Appearing Attack (AA), we proposed the nested-AE, which combines two AEs together to attack object detectors in both long and short distance. We also designed diverse styles of AEs to make AA more surreptitious. Evaluation results show that our AEs can attack the state-of-the-art real-time object detectors (i.e., YOLO V3 and faster-RCNN) at the success rate up to 92.4% with varying distance from 1m to 25m and angles from -60° to 60°. Our AEs are also demonstrated to be highly transferable, capable of attacking another three state-of-the-art black-box models with high success rate.

cs.CV↗

arXiv2019

Elastic scattering from rough surfaces in three dimensions

Consider the elastic scattering of a plane or point incident wave by an unbounded and rigid rough surface. The angular spectrum representation (ASR) for the time-harmonic Navier equation is derived in three dimensions. The ASR is utilized as a radiation condition to the elastic rough surface scattering problem. The uniqueness is proved through a Rellich-type identity for surfaces given by uniformly Lipschitz functions. In the case of flat surfaces with a local perturbation, we deduce an equivalent variational formulation in a truncated bounded domain and show the existence results for general incoming waves. The main ingredient of the proof is the radiating behavior of the Green tensor to the first boundary value problem of the Navier equation in a half space.

math.AP↗

arXiv2019

Geographically Coordinated Primary Frequency Control Technical Report

Primary Frequency Control (PFC) is a fast acting mechanism used to ensure high-quality power for the grid that is becoming an increasingly attractive option for load participation. Due to speed requirement and other considerations, it is often desirable to have distributed control laws. Current distributed PFC designs assume that the costs at each geographic location are independent. However, many networked systems, such as those for cloud computing, have interdependent costs across locations and therefore need geographic coordination. In this paper, distributed control laws are designed for interdependent, geo-distributed loads in PFC based on the optimality conditions of the global system. The controlled frequencies are provably stable, and the final equilibrium point is proven to strike an optimal balance between load participation and the frequency's deviation from its nominal set point. The case with a significant communication delay for control is also analyzed. We evaluate the proposed control laws with realistic numerical simulations. Under current technology, the proposed control laws achieve a convergence time that is smaller than droop control alone and is comparable to that of distributed control without interdependent costs. Results also highlight significant cost savings over existing approaches under a variety of settings.

math.OC↗

arXiv2019

Soft excess in the quiescent Be/X-ray pulsar RX J0812.4-3114

We report a 72 ks XMM-Newton observation of the Be/X-ray pulsar (BeXRP) RX J0812.4-3114 in quiescence ($L_X \approx 1.6 \times 10^{33}~\mathrm{erg~s^{-1}}$). Intriguingly, we find a two component spectrum, with a hard power-law ($Γ\approx 1.5$) and a soft blackbody-like excess below $\approx 1~\mathrm{keV}$. The blackbody component is consistent in $kT$ with a prior quiescent Chandra observation reported by Tsygankov et al. and has an inferred blackbody radius of $\approx 10~\mathrm{km}$, consistent with emission from the entire neutron star (NS) surface. There is also mild evidence for an absorption line at $\approx 1~\mathrm{keV}$ and/or $\approx 1.4~\mathrm{keV}$. The hard component shows pulsations at $P \approx 31.908~\mathrm{s}$ (pulsed fraction $0.84 \pm 0.10$), agreeing with the pulse period seen previously in outbursts, but no pulsations were found in the soft excess (pulsed fraction $\lesssim 31\%$). We conclude that the pulsed hard component suggests low-level accretion onto the neutron star poles, while the soft excess seems to originate from the entire NS surface. We speculate that, in quiescence, the source switches between a soft thermal-dominated state (when the propeller effect is at work) and a relatively hard state with low-level accretion, and use the propeller cutoff to estimate the magnetic field of the system to be $\lesssim 8.4 \times 10^{11}~\mathrm{G}$. We compare the quiescent thermal $L_X$ predicted by the standard deep crustal heating model to our observations and find that RX J0812.4-3114 has a high thermal $L_X$, at or above the prediction for minimum cooling mechanisms. This suggests that RX J0812.4-3114 either contains a relatively low-mass NS with minimum cooling, or that the system may be young enough that the NS has not fully cooled from the supernova explosion.

astro-ph.HE↗

arXiv2019

Uniqueness to some inverse source problems for the wave equation in unbounded domains

This paper is concerned with inverse acoustic source problems in an unbounded domain with dynamical boundary surface data of Dirichlet kind. The measurement data are taken at a surface far away from the source support. We prove uniqueness in recovering source terms of the form $f(x)g(t)$ and $f(x_1,x_2,t) h(x_3)$, where $g(t)$ and $h(x_3)$ are given and $x=(x_1, x_2, x_3)$ is the spatial variable in three dimensions. Without these a priori information, we prove that the boundary data of a family of solutions can be used to recover general source terms depending on both time and spatial variables. For moving point sources radiating periodic signals, the data recorded at four receivers are proven sufficient to uniquely recover the orbit function. Simultaneous determination of embedded obstacles and source terms was verified in an inhomogeneous background medium using the observation data of infinite time period. Our approach depends heavily on the Laplace transform.

math.AP↗

arXiv2019

A Learning-to-Infer Method for Real-Time Power Grid Multi-Line Outage Identification

Identifying a potentially large number of simultaneous line outages in power transmission networks in real time is a computationally hard problem. This is because the number of hypotheses grows exponentially with the network size. A new "Learning-to-Infer" method is developed for efficient inference of every line status in the network. Optimizing the line outage detector is transformed to and solved as a discriminative learning problem based on Monte Carlo samples generated with power flow simulations. A major advantage of the developed Learning-to-Infer method is that the labeled data used for training can be generated in an arbitrarily large amount rapidly and at very little cost. As a result, the power of offline training is fully exploited to learn very complex classifiers for effective real-time multi-line outage identification. The proposed methods are evaluated in the IEEE 30, 118 and 300 bus systems. Excellent performance in identifying multi-line outages in real time is achieved with a reasonably small amount of data.

cs.LG↗

arXiv2019

PyOD: A Python Toolbox for Scalable Outlier Detection

PyOD is an open-source Python toolbox for performing scalable outlier detection on multivariate data. Uniquely, it provides access to a wide range of outlier detection algorithms, including established outlier ensembles and more recent neural network-based approaches, under a single, well-documented API designed for use by both practitioners and researchers. With robustness and scalability in mind, best practices such as unit testing, continuous integration, code coverage, maintainability checks, interactive examples and parallelization are emphasized as core components in the toolbox's development. PyOD is compatible with both Python 2 and 3 and can be installed through Python Package Index (PyPI) or https://github.com/yzhao062/pyod.

cs.LG↗

arXiv2019

Detecting dark photon dark matter with Gaia-like astrometry observations

A class of dark photon dark matter models with ultralight masses would lead to oscillation of a test body through a coupling with baryons or $B-L$ charge. This periodical oscillation of an observer results in swing of a star's apparent position due to the effect of aberration of light, which could be probed with high-precision astrometry observations of stars in the Milky Way. We propose to use the observations of stellar positions of a number of stars by Gaia to search for this kind of dark photon dark matter. We show that this astrometry method is able to give promising sensitivities to search for the dark photon dark matter in the mass range of $10^{-23}\sim10^{-21}$ eV.

hep-ph↗

arXiv2019

Dark Photon Dark Matter Produced by Axion Oscillations

Despite growing interest and extensive effort to search for ultralight dark matter in the form of a hypothetical dark photon, how it fits into a consistent cosmology is unclear. Several dark photon dark matter production mechanisms proposed previously are known to have limitations, at least in certain mass regimes of experimental interest. In this letter, we explore a novel mechanism, where a coherently oscillating axion-like field can efficiently transfer its energy density to a dark photon field via a tachyonic instability. The residual axion relic is subsequently depleted via couplings to the visible sector, leaving only the dark photon as dark matter. We ensure that the cosmologies of both the axion and dark photon are consistent with existing constraints. We find that the mechanism works for a broad range of dark photon masses, including those of interest for ongoing experiments and proposed detection techniques.

hep-ph↗

arXiv2019

Enumeration on row-increasing tableaux of shape $2 \times n$

Recently O. Pechenik studied the cyclic sieving of increasing tableaux of shape $2\times n$, and obtained a polynomial on the major index of these tableaux, which is a $q$-analogue of refined small Schröder numbers. We define row-increasing tableaux and study the major index and amajor index of row-increasing tableaux of shape $2 \times n$. The resulting polynomials are both $q$-analogues of refined large Schröder numbers. For both results we give bijective proofs.

math.CO↗