H. Xu - LM-Orsay

H. Xu
Are you H. Xu?

Claim your profile, edit publications, add additional information:

Contact Details

H. Xu

Pubs By Year

Pub Categories

Computer Science - Learning (8)
Physics - Mesoscopic Systems and Quantum Hall Effect (7)
Statistics - Machine Learning (5)
Instrumentation and Methods for Astrophysics (4)
Quantum Physics (4)
Physics - Optics (4)
Computer Science - Computer Vision and Pattern Recognition (4)
Physics - Plasma Physics (3)
Computer Science - Computation and Language (3)
Physics - Materials Science (3)
Earth and Planetary Astrophysics (2)
Solar and Stellar Astrophysics (2)
Computer Science - Networking and Internet Architecture (2)
Physics - Instrumentation and Detectors (2)
Mathematics - Probability (2)
Computer Science - Artificial Intelligence (2)
Astrophysics of Galaxies (1)
Nonlinear Sciences - Pattern Formation and Solitons (1)
Nuclear Theory (1)
Mathematics - Optimization and Control (1)
Nuclear Experiment (1)
Computer Science - Information Theory (1)
Mathematics - Information Theory (1)
Statistics - Methodology (1)
Computer Science - Sound (1)
Quantitative Biology - Quantitative Methods (1)
Nonlinear Sciences - Chaotic Dynamics (1)
High Energy Physics - Phenomenology (1)
Statistics - Theory (1)
Computer Science - Computer Science and Game Theory (1)
Computer Science - Cryptography and Security (1)
Mathematics - Numerical Analysis (1)
Mathematics - Analysis of PDEs (1)
Physics - Biological Physics (1)
Quantitative Biology - Cell Behavior (1)
Physics - Data Analysis; Statistics and Probability (1)
Mathematics - Statistics (1)
Mathematics - Combinatorics (1)
Computer Science - Distributed; Parallel; and Cluster Computing (1)
Computer Science - Robotics (1)
Computer Science - Multiagent Systems (1)
Physics - Medical Physics (1)
Computer Science - Human-Computer Interaction (1)

Publications Authored By H. Xu

In this paper, we discuss the existence of solutions for the following non-local critical systems: \begin{equation*} \begin{cases} (-\Delta)^{s}u= \mu_{1}|u|^{2^{\ast}-2}u+\frac{\alpha\gamma}{2^{\ast}}|u|^{\alpha-2}u|v|^{\beta} \ \ \ in \ \ R^{N}, (-\Delta)^{s}v= \mu_{2}|v|^{2^{\ast}-2}v+\frac{\beta\gamma}{2^{\ast}}|u|^{\alpha}|v|^{\beta-2}v\ in \ \ R^{N}, u,v\in D_{s}(R^{N}). \end{cases} \end{equation*} By using the Nehari manifold,\ under proper conditions, we establish the existence and nonexistence of positive least energy solution. Read More

A sperm-driven micromotor is presented as cargo-delivery system for the treatment of gynecological cancers. This particular hybrid micromotor is appealing to treat diseases in the female reproductive tract, the physiological environment that sperm cells are naturally adapted to swim in. Here, the single sperm cell serves as an active drug carrier and as driving force, taking advantage of its swimming capability, while a laser-printed microstructure coated with a nanometric layer of iron is used to guide and release the sperm in the desired area by an external magnet and structurally imposed mechanical actuation, respectively. Read More

We address the problem of activity detection in continuous, untrimmed video streams. This is a difficult task that requires extracting meaningful spatio-temporal features to capture activities, accurately localizing the start and end times of each activity, and also dealing with very large data volumes. We introduce a new model, Region Convolutional 3D Network (R-C3D), which encodes the video streams using a three-dimensional fully convolutional network, then generates candidate temporal regions containing activities, and finally classifies selected regions into specific activities. Read More

We propose the first multistage intervention framework that tackles fake news in social networks by combining reinforcement learning with a point process network activity model. The spread of fake news and mitigation events within the network is modeled by a multivariate Hawkes process with additional exogenous control terms. By choosing a feature representation of states, defining mitigation actions and constructing reward functions to measure the effectiveness of mitigation activities, we map the problem of fake news mitigation into the reinforcement learning framework. Read More

Non-Hermitian systems exhibit phenomena that are qualitatively different from those of Hermitian systems and have been exploited to achieve a number of ends, including the generation of exceptional points, nonreciprocal dynamics, non-orthogonal normal modes, and topological operations. However to date these effects have only been accessible with nearly-degenerate modes (i.e. Read More

Segment routing is an emerging technology to simplify traffic engineering implementation in WANs. It expresses an end-to-end logical path as a sequence of segments, each of which is represented by a middlepoint. In this paper, we arguably conduct the first systematic study of traffic engineering with segment routing in SDN based WANs. Read More

In this paper, we identify and study a fundamental, yet underexplored, phenomenon in security games, which we term the Curse of Correlation (CoC). Specifically, we observe that there is inevitable correlation among the protection status of different targets. Such correlation is a crucial concern, especially in spatio-temporal domains like conservation area patrolling, where attackers can monitor patrollers at certain areas and then infer their patrolling routes using such correlation. Read More

We provide a hybrid method that captures the polynomial speed of convergence and polynomial speed of mixing for Markov processes. The hybrid method that we introduce is based on the coupling technique and renewal theory. We propose to replace some estimates in classical results about the ergodicity of Markov processes by numerical simulations when the corresponding analytical proof is difficult. Read More

We report fabrication and measurement of a device where closely-placed two parallel InAs nanowires (NWs) are contacted by source and drain normal metal electrodes. Established technique includes selective deposition of double nanowires onto a previously defined gate region. By tuning the junction with the finger bottom gates, we confirmed the formation of parallel double quantum dots, one in each NW, with a finite electrostatic coupling between each other. Read More

The study on point sources in astronomical images is of special importance, since most energetic celestial objects in the Universe exhibit a point-like appearance. An approach to recognize the point sources (PS) in the X-ray astronomical images using our newly designed granular binary-tree support vector machine (GBT-SVM) classifier is proposed. First, all potential point sources are located by peak detection on the image. Read More

In this paper, the calorimetric power measurement method for electron cyclotron resonance heating system on EAST are presented. This method requires measurement of the water flow through the cooling circuits and the input and output water temperatures in each cooling circuit. Usually, the inlet water temperature is controlled to be stable to get more accurate results. Read More

We present stellar properties (mass, age, radius, distances) of 57 stars from a seismic inference using full-length data sets from Kepler. These stars comprise active stars, planet-hosts, solar-analogs, and binary systems. We validate the distances derived from the astrometric Gaia-Tycho solution. Read More

The gyrotron is the most important device in the ECRH system. The cathode power supply is one of the most important ancillary devices for gyrotron. Some interesting transient phenomena about the cathode voltage and the cathode current was found in the gyrotron operation. Read More

The Allan Variance (AV) is a widely used quantity in areas focusing on error measurement as well as in the general analysis of variance for autocorrelated processes in domains such as engineering and, more specifically, metrology. The form of this quantity is widely used to detect noise patterns and indications of stability within signals. However, the properties of this quantity are not known for commonly occurring processes whose covariance structure is non-stationary and, in these cases, an erroneous interpretation of the AV could lead to misleading conclusions. Read More

Many real-world applications require robust algorithms to learn point process models based on a type of incomplete data --- the so-called short doubly-censored (SDC) event sequences. In this paper, we study this critical problem of quantitative asynchronous event sequence analysis under the framework of Hawkes processes by leveraging the general idea of data synthesis. In particular, given SDC event sequences observed in a variety of time intervals, we propose a sampling-stitching data synthesis method --- sampling predecessor and successor for each SDC event sequence from potential candidates and stitching them together to synthesize long training sequences. Read More

SAGA is a fast incremental gradient method on the finite sum problem and its effectiveness has been tested on a vast of applications. In this paper, we analyze SAGA on a class of non-strongly convex and non-convex statistical problem such as Lasso, group Lasso, Logistic regression with $\ell_1$ regularization, linear regression with SCAD regularization and Correct Lasso. We prove that SAGA enjoys the linear convergence rate up to the statistical estimation accuracy, under the assumption of restricted strong convexity (RSC). Read More

Multi-agent path finding (MAPF) is well-studied in artificial intelligence, robotics, theoretical computer science and operations research. We discuss issues that arise when generalizing MAPF methods to real-world scenarios and four research directions that address them. We emphasize the importance of addressing these issues as opposed to developing faster methods for the standard formulation of the MAPF problem. Read More

Although deep learning has yielded impressive performance for face recognition, many studies have shown that different networks learn different feature maps: while some networks are more receptive to pose and illumination others appear to capture more local information. Thus, in this work, we propose a deep heterogeneous feature fusion network to exploit the complementary information present in features generated by different deep convolutional neural networks (DCNNs) for template-based face recognition, where a template refers to a set of still face images or video frames from different sources which introduces more blur, pose, illumination and other variations than traditional face datasets. The proposed approach efficiently fuses the discriminative information of different deep features by 1) jointly learning the non-linear high-dimensional projection of the deep features and 2) generating a more discriminative template representation which preserves the inherent geometry of the deep features in the feature space. Read More

In software-defined networking (SDN), as data plane scale expands, scalability and reliability of the control plane has become major concerns. To mitigate such concerns, two kinds of solutions have been proposed separately. One is multi-controller architecture, i. Read More

Consider a coloring of a graph such that each vertex is assigned a fraction of each color, with the total amount of colors at each vertex summing to $1$. We define the fractional defect of a vertex $v$ to be the sum of the overlaps with each neighbor of $v$, and the fractional defect of the graph to be the maximum of the defects over all vertices. Note that this coincides with the usual definition of defect if every vertex is monochromatic. Read More

We propose an effective method to solve the event sequence clustering problems based on a novel Dirichlet mixture model of a special but significant type of point processes --- Hawkes process. In this model, each event sequence belonging to a cluster is generated via the same Hawkes process with specific parameters, and different clusters correspond to different Hawkes processes. The prior distribution of the Hawkes processes is controlled via a Dirichlet process. Read More

Topological optical states exhibit unique immunity to defects and the ability to propagate without losses rendering them ideal for photonic applications.A powerful class of such states is based on time-reversal symmetry breaking of the optical response.However, existing proposals either involve sophisticated and bulky structural designs or can only operate in the microwave regime. Read More

Metal-Insulator-Metal tunnel junctions (MIMTJ) are common throughout the microelectronics industry. The industry standard AlOx tunnel barrier, formed through oxygen diffusion into an Al wetting layer, is plagued by internal defects and pinholes which prevent the realization of atomically-thin barriers demanded for enhanced quantum coherence. In this work, we employed in situ scanning tunneling spectroscopy (STS) along with molecular dynamics simulations to understand and control the growth of atomically thin Al2O3 tunnel barriers using atomic layer deposition (ALD). Read More

In this paper, we consider stochastic dual coordinate (SDCA) {\em without} strongly convex assumption or convex assumption. We show that SDCA converges linearly under mild conditions termed restricted strong convexity. This covers a wide array of popular statistical models including Lasso, group Lasso, and logistic regression with $\ell_1$ regularization, corrected Lasso and linear regression with SCAD regularizer. Read More

Ensemble learning has been widely employed by mobile applications, ranging from environmental sensing to activity recognitions. One of the fundamental issue in ensemble learning is the trade-off between classification accuracy and computational costs, which is the goal of ensemble pruning. During crowdsourcing, the centralized aggregator releases ensemble learning models to a large number of mobile participants for task evaluation or as the crowdsourcing learning results, while different participants may seek for different levels of the accuracy-cost trade-off. Read More

This study aims to evaluate and compare the climatic drivers of Qinghai spruce, a species endemic to northwest China, over an elevational gradient, from its lower distributional margin to its upper distributional margin. We sampled pure Qinghai spruce stands and obtained time-series of growth from 25 trees each in three populations across it distributional range in the Helan Mountains. We found that trees at the lower distributional margin have lower first-order autocorrelation coefficients in annual growth and have experienced an increase in locally-absent rings (stem-growth cessation) since 2001. Read More

We explore signatures of the non-Markovianity in the time-resolved energy transfer processes for quantum open systems. Focusing on typical systems such as the exact solvable damped Jaynes-Cummings model and the general spin-boson model, we establish quantitative links between the time-resolved energy current and the symmetric logarithmic derivative quantum Fisher information (SLD-QFI) flow, one of measures quantifying the non-Markovianity, within the framework of non-Markovian master equations in time-local forms. From the relationships, we find in the damped Jaynes-Cummings model that the SLD-QFI backflow from the reservoir to the system always correlates with an energy backflow, thus we can directly witness the non-Markovianity from the dynamics of the energy current. Read More

In this work, we provide two complementary perspectives for the (spectral) stability of traveling waves in Hamiltonian nonlinear dynamical lattices, of which the Fermi-Pasta-Ulam and the Toda lattice are prototypical examples. One is as an eigenvalue problem for a stationary solution in a co-traveling frame, while the other is as a periodic orbit modulo shifts. We connect the eigenvalues of the former with the Floquet multipliers of the latter and based on this formulation derive an energy-based spectral stability criterion. Read More

It is shown by particle-in-cell simulation that intense circularly polarized (CP) laser light can be contained in the cavity of a solid-density circular Al-plasma shell for hundreds of light-wave periods before it is dissipated by laser-plasma interaction. A right-hand CP laser pulse can propagate almost without reflection into the cavity through a highly magnetized overdense H-plasma slab filling the entrance hole. The entrapped laser light is then multiply reflected at the inner surfaces of the slab and shell plasmas, gradually losing energy to the latter. Read More

(submitted to MNRAS January 10, 2017) We present synthetic observations for the first generations of galaxies in the Universe and make predictions for future deep field observations for redshifts greater than 6. Due to the strong impact of nebular emission lines and the relatively compact scale of HII regions, high resolution cosmological simulations and a robust suite of analysis tools are required to properly simulate spectra. We created a software pipeline consisting of FSPS, Hyperion, Cloudy and our own tools to generate synthetic IR observations from a fully three-dimensional arrangement of gas, dust, and stars. Read More

We consider large-scale Markov decision processes (MDPs) with a risk measure of variability in cost, under the risk-aware MDPs paradigm. Previous studies showed that risk-aware MDPs, based on a minimax approach to handling risk, can be solved using dynamic programming for small to medium sized problems. However, due to the "curse of dimensionality", MDPs that model real-life problems are typically prohibitively large for such approach. Read More

We consider the problem of learning from noisy data in practical settings where the size of data is too large to store on a single machine. More challenging, the data coming from the wild may contain malicious outliers. To address the scalability and robustness issues, we present an online robust learning (ORL) approach. Read More

Hybrid plasmonic lasers provide deep subwavelength optical confinement, strongly enhanced light-matter interaction and together with nanoscale footprint promise new applications in optical communication, bio-sensing and photolithography. The subwavelength hybrid plasmonic lasers reported so far often use bottom up grown nanowires, nanorods and nanosquares, making it difficult to integrate these devices into industry-relevant high density plasmonic circuits. Here, we report the first experimental demonstration of AlGaInP based, red-emitting hybrid plasmonic lasers at room temperature using lithography based fabrication processes. Read More

Hybrid superconductor-semiconducting nanowire devices provide an ideal platform to investigating novel intragap bound states, such as the Andreev bound states (ABSs), Yu-Shiba-Rusinov (YSR) states, and the Majorana bound states. The competition between Kondo correlations and superconductivity in Josephson quantum dot (QD) devices results in two different ground states and the occurrence of a 0-$\pi$ quantum phase transition. Here we report on transport measurements on hybrid superconductor-InSb nanowire QD devices with different device geometries. Read More

The Kepler space telescope yielded unprecedented data for the study of solar-like oscillations in other stars. The large samples of multi-year observations posed an enormous data analysis challenge that has only recently been surmounted. Asteroseismic modeling has become more sophisticated over time, with better methods gradually developing alongside the extended observations and improved data analysis techniques. Read More

Zitterbewegung (ZB) is a phenomenon in relativistic quantum systems where the electron wave packet exhibits a trembling or oscillating behavior during its motion, caused by its interaction or coupling with the negative energy state. To directly observe ZB in electronic systems is difficult, due to the challenges associated with the atomic scale wavelength of the electron. Photonic systems offer an alternative paradigm. Read More

One of the key tasks of sentiment analysis of product reviews is to extract product aspects or features that users have expressed opinions on. In this work, we focus on using supervised sequence labeling as the base approach to performing the task. Although several extraction methods using sequence labeling methods such as Conditional Random Fields (CRF) and Hidden Markov Models (HMM) have been proposed, we show that this supervised approach can be significantly improved by exploiting the idea of concept sharing across multiple domains. Read More

We uncover a superscattering behavior of pseudospin-1 wave from weak scatterers in the subwavelength regime where the scatterer size is much smaller than wavelength. The phenomenon manifests itself as unusually strong scattering characterized by extraordinarily large values of the cross section even for arbitrarily weak scatterer strength. We establish analytically and numerically that the physical origin of superscattering is revival resonances, for which the conventional Born theory breaks down. Read More

The effects of volume corrections and resonance decays (the resulting correlations between positive charges and negative charges) on cumulants of net-proton distributions and net-charge distributions are investigated by using a Monte Carlo hadron resonance gas ({\tt MCHRG}) model. The required volume distributions are generated by a Monte Carlo Glauber ({\tt MC-Glb}) model. Except the variances of net-charge distributions, the {\tt MCHRG} model with more realistic simulations of volume corrections, resonance decays and acceptance cuts can reasonably explain the data of cumulants of net-proton distributions and net-charge distributions reported by the STAR collaboration. Read More

State-of-the-art i-vector based speaker verification relies on variants of Probabilistic Linear Discriminant Analysis (PLDA) for discriminant analysis. We are mainly motivated by the recent work of the joint Bayesian (JB) method, which is originally proposed for discriminant analysis in face verification. We apply JB to speaker verification and make three contributions beyond the original JB. Read More

Product Community Question Answering (PCQA) provides useful information about products and their features (aspects) that may not be well addressed by product descriptions and reviews. We observe that a product's compatibility issues with other products are frequently discussed in PCQA and such issues are more frequently addressed in accessories, i.e. Read More

A crucial result in quantum chaos, which has been established for a long time, is that the spectral properties of classically integrable systems generically are described by Poisson statistics whereas those of time-reversal symmetric, classically chaotic systems coincide with those of random matrices from the Gaussian orthogonal ensemble (GOE). Does this result hold for two-dimensional Dirac material systems? To address this fundamen- tal question, we investigate the spectral properties in a representative class of graphene billiards with shapes of classically integrable circular-sector billiards. Naively one may expect to observe Poisson statistics, which is indeed true for energies close to the band edges where the quasiparticle obeys the Schr\"odinger equation. Read More

Gamma-ray polarimetry is a new powerful tool to study the processes responsible for the emission from astrophysical sources and the environments in which this emission takes place. Few successful polarimetric measurements have however been performed thus far in the gamma-ray energy band due to the difficulties involved. POLAR is a dedicated polarimeter designed to perform high precision measurements of the polarization of the emission from gamma-ray burst in the 50-500 keV energy range. Read More

Signatures of Majorana fermion bound states in one-dimensional topological insulator (TI) nanowires with proximity effect induced superconductivity are studied. The phase diagram and energy spectra are calculated for single TI nanowires and it is shown that the nanowires can be in the topological invariant phases of winding numbers $W=0, \pm 1$, and $\pm 2$ corresponding to the cases with zero, one and two pairs of Majorana fermions in the single TI nanowires. It is also shown that the topological winding numbers, i. Read More

Research on peer effects in sociology has been focused for long on social influence power to investigate the social foundations for social interactions. This paper extends Xu(2011)'s large--network--based game model by allowing for social-influence-dependent peer effects. In a large network, we use the Katz--Bonacich centrality to measure individuals' social influences. Read More

Black TiO2 nanoparticles with a crystalline-core and amorphous-shell structure exhibit superior optoelectronic properties in comparison with pristine TiO2. The fundamental mechanisms underlying these enhancements, however, remain unclear, largely due to the inherent complexities and limitations of powder materials. Here, we fabricate TiO2 homojunction films consisting of an oxygen-deficient amorphous layer on top of a highly crystalline layer, to simulate the structural/functional configuration of black TiO2 nanoparticles. Read More

An organic-inorganic halide perovskite of CH3NH3SnI3 with significantly improved structural stability is obtained via pressure-induced amorphization and recrystallization. In situ high-pressure resistance measurements reveal an increased electrical conductivity by 300% in the pressure-treated perovskite. Photocurrent measurements also reveal a substantial enhancement in visible-light responsiveness. Read More

Robust perception-action models should be learned from training data with diverse visual appearances and realistic behaviors, yet current approaches to deep visuomotor policy learning have been generally limited to in-situ models learned from a single vehicle or a simulation environment. We advocate learning a generic vehicle motion model from large scale crowd-sourced video data, and develop an end-to-end trainable architecture for learning to predict a distribution over future vehicle egomotion from instantaneous monocular camera observations and previous vehicle state. Our model incorporates a novel FCN-LSTM architecture, which can be learned from large-scale crowd-sourced vehicle action data, and leverages available scene segmentation side tasks to improve performance under a privileged learning paradigm. Read More

Product reviews contain a lot of useful information about product features and customer opinions. One important product feature is the complementary entity (products) that may potentially work together with the reviewed product. Knowing complementary entities of the reviewed product is very important because customers want to buy compatible products and avoid incompatible ones. Read More

In this paper, we give a notation on the Singleton bounds for linear codes over a finite commutative quasi-Frobenius ring in the work of Shiromoto [5]. We show that there exists a class of finite commutative quasi-Frobenius rings. The Singleton bounds for linear codes over such rings satisfy \[ \frac{d(C)-1}{A}\leq n-\log_{|R|}|C|. Read More