arXiv subjects

Jun Xu

Publications and source records attributed to Jun Xu.

At least 451 records · Page 25Linked to original sources

arXiv2017

Spherical Paragraph Model

Representing texts as fixed-length vectors is central to many language processing tasks. Most traditional methods build text representations based on the simple Bag-of-Words (BoW) representation, which loses the rich semantic relations between words. Recent advances in natural language processing have shown that semantically meaningful representations of words can be efficiently acquired by distributed models, making it possible to build text representations based on a better foundation called the Bag-of-Word-Embedding (BoWE) representation. However, existing text representation methods using BoWE often lack sound probabilistic foundations or cannot well capture the semantic relatedness encoded in word vectors. To address these problems, we introduce the Spherical Paragraph Model (SPM), a probabilistic generative model based on BoWE, for text representation. SPM has good probabilistic interpretability and can fully leverage the rich semantics of words, the word co-occurrence information as well as the corpus-wide information to help the representation learning of texts. Experimental results on topical classification and sentiment analysis demonstrate that SPM can achieve new state-of-the-art performances on several benchmark datasets.

cs.CL↗

arXiv2017

Diode-pumped mode-locked femtosecond 2-um Tm:CaYAlO4 laser

Continuous-wave mode-locked femtosecond 2 um solid-state laser with a c-cut Tm:CaYAlO4 as gain medium was experimentally demonstrated. The mode locked laser generated stable pulses with average output power as high as 531 mW, pulse duration of 496 fs, and repetition rate of 97 MHz at 1975 nm. The research results show that Tm:CaYAlO4 is an excellent gain medium for femtosecond pulse generation at 2um wavelength.

physics.optics↗

arXiv2017

From Physical to Cyber: Escalating Protection for Personalized Auto Insurance

Nowadays, auto insurance companies set personalized insurance rate based on data gathered directly from their customers' cars. In this paper, we show such a personalized insurance mechanism -- wildly adopted by many auto insurance companies -- is vulnerable to exploit. In particular, we demonstrate that an adversary can leverage off-the-shelf hardware to manipulate the data to the device that collects drivers' habits for insurance rate customization and obtain a fraudulent insurance discount. In response to this type of attack, we also propose a defense mechanism that escalates the protection for insurers' data collection. The main idea of this mechanism is to augment the insurer's data collection device with the ability to gather unforgeable data acquired from the physical world, and then leverage these data to identify manipulated data points. Our defense mechanism leveraged a statistical model built on unmanipulated data and is robust to manipulation methods that are not foreseen previously. We have implemented this defense mechanism as a proof-of-concept prototype and tested its effectiveness in the real world. Our evaluation shows that our defense mechanism exhibits a false positive rate of 0.032 and a false negative rate of 0.013.

cs.CR↗

arXiv2017

Chemical freeze-out in relativistic heavy-ion collisions

One surprising result in relativistic heavy-ion collisions is that the abundance of various particles measured in experiments is consistent with the picture that they reach chemical equilibrium at a temperature much higher than the temperature they freeze out kinetically. Using a multiphase transport model to study particle production in these collisions, we find that the above result is due to the constancy of the entropy per particle during the evolution of the hadronic matter from the chemical to the kinetic freeze-out. We further use a hadron resonance gas model to illustrate the result from the transport model study.

nucl-th↗

arXiv2017

Spectral indices for radio emission of 228 pulsars

We determine spectral indices of 228 pulsars by using Parkes pulsar data observed at 1.4 GHz, among which 200 spectra are newly determined. The indices are distributed in the range from -4.84 to -0.46.Together with known pulsar spectra from literature, we tried to find clues to the pulsar emission process. The weak correlations between the spectral index, the spin-down energy loss rate $\dot{E}$ and the potential drop in the polar gap $ΔΨ$ hint that emission properties are related to the particle acceleration process in a pulsar's magnetosphere.

astro-ph.HE↗

arXiv2017

Constraining simultaneously nuclear symmetry energy and neutron-proton effective mass splitting with nucleus giant resonances from a dynamical approach

With a newly improved isospin- and momentum-dependent interaction and an isospin-dependent Boltzmann-Uehling-Uhlenbeck transport model, we have investigated the effects of the slope parameter $L$ of the nuclear symmetry energy and the isospin splitting of the nucleon effective mass $m_{n-p}^*=(m_n^*-m_p^*)/m$ on the centroid energy of the isovector giant dipole resonance and the electric dipole polarizability in $^{208}$Pb. With the isoscalar nucleon effective mass $m_s^*=0.7m$ constrained by the empirical optical potential, we obtain a constraint of $L=64.29\pm11.84 (\rm MeV)$ and $m_{n-p}^*= (-0.019 \pm 0.090)δ$, with $δ$ being the isospin asymmetry of nuclear medium. With the isoscalar nucleon effective mass $m_s^*=0.84m$ extracted from the excitation energy of the isoscalar giant quadruple resonance in $^{208}$Pb, we obtain a constraint of $L=53.85\pm10.29 (\rm MeV)$ and $m_{n-p}^*= (0.216 \pm 0.114)δ$.

nucl-th↗

arXiv2017

Mechanical Wave Propagation within Nanogold Granular Crystals

We computationally investigate the wave propagation characteristics of nanoscopic granular crystals composed of one-dimensionally arrayed gold nanoparticles using molecular dynamics simulation. We examine two basic configurations, i.e. homogeneous lattices and diatomic lattices with mass-mismatch. We discover that homogeneous lattices of gold nanospheres support weakly dissipative and highly localized solitary wave at 300 K, while diatomic lattices have a good tuning ability of transmittance and wave speed. We establish a validated nonlinear spring contact model with the consideration of complex interactions between gold nanospheres which reveals the physical nature of wave behaviors at nanoscale. This work sheds light on the application of nanogold as a novel mechanical wave tuner, qualitatively and fundamentally different from its counterpart granular materials at meso- and macroscale.

cond-mat.mes-hall↗

arXiv2017

Long-range transverse Ising model built with dipolar condensates in two-well arrays

Dipolar Bose-Einstein condensates in an array of double-well potentials realize an effective transverse Ising model with peculiar inter-layer interactions, that may result under proper conditions in an anomalous first-order ferromagnetic-antiferromagnetic phase transition, and nontrivial phases due to frustration. The considered setup as well allows the study of Kibble-Zurek defect formation, whose kink statistics follows that expected from the universality class of the mean-field transverse Ising model in 1D. Furthermore, random occupation of each layer of the stack leads to random effective Ising interactions and generation of local transverse fields, thus allowing the study of Anderson-like localization of imbalance perturbations in the two-well stack under controllable conditions.

cond-mat.quant-gas↗

arXiv2016

Internet of Things Applications: Animal Monitoring with Unmanned Aerial Vehicle

In animal monitoring applications, both animal detection and their movement prediction are major tasks. While a variety of animal monitoring strategies exist, most of them rely on mounting devices. However, in real world, it is difficult to find these animals and install mounting devices. In this paper, we propose an animal monitoring application by utilizing wireless sensor networks (WSNs) and unmanned aerial vehicle (UAV). The objective of the application is to detect locations of endangered species in large-scale wildlife areas and monitor movement of animals without any attached devices. In this application, sensors deployed throughout the observation area are responsible for gathering animal information. The UAV flies above the observation area and collects the information from sensors. To achieve the information efficiently, we propose a path planning approach for the UAV based on a Markov decision process (MDP) model. The UAV receives a certain amount of reward from an area if some animals are detected at that location. We solve the MDP using Q-learning such that the UAV prefers going to those areas that animals are detected before. Meanwhile, the UAV explores other areas as well to cover the entire network and detects changes in the animal positions. We first define the mathematical model underlying the animal monitoring problem in terms of the value of information (VoI) and rewards. We propose a network model including clusters of sensor nodes and a single UAV that acts as a mobile sink and visits the clusters. Then, one MDP-based path planning approach is designed to maximize the VoI while reducing message delays. The effectiveness of the proposed approach is evaluated using two real-world movement datasets of zebras and leopard. Simulation results show that our approach outperforms greedy, random heuristics and the path planning based on the traveling salesman problem.

cs.AI↗

arXiv2016

Evidence for topological proximity effect in graphene coupled to topological insulator

The emergence of topological order in graphene is in great demand for the realization of quantum spin Hall states. Recently, it is theoretically proposed that the spin textures of surface states in topological insulator can be directly transferred to graphene by means of proximity effect. Here we report the observations of the topological proximity effect in the graphene-topological insulator Bi2Se3 heterojunctions via magnetotransport measurements. The coupling between the p_z orbitals of graphene and the p orbitals of surface states on the Bi2Se3 bottom surface can be enhanced by applying perpendicular negative magnetic field, resulting in a giant negative magnetoresistance at the Dirac point up to about -91%. An obvious resistivity dip in the transfer curve at the Dirac point is also observed in the hybrid devices, which is consistent with the theoretical predictions of the distorted Dirac bands with unique spin textures inherited from Bi2Se3 surface states.

cond-mat.mes-hall↗

arXiv2016

Collision energy dependence of elliptic flow splitting between particles and their antiparticles from an extended multiphase transport model

Based on an extended multiphase transport model, which includes mean-field potentials in both the partonic and hadronic phases, uses the mix-event coalescence, and respects charge conservation during the hadronic evolution, we have studied the collision energy dependence of the elliptic flow splitting between particles and their antiparticles. This extended transport model reproduces reasonably well the experimental data at lower collision energies but only describes qualitatively the elliptic flow splitting at higher beam energies. The present study thus indicates the existence of other mechanisms for the elliptic flow splitting besides the mean-field potentials and the need of further improvements of the multiphase transport model.

nucl-th↗

arXiv2016

Universal spatiotemporal dynamics of spontaneous superfluidity breakdown in the presence of synthetic gauge fields

According to the famous Kibble-Zurek mechanism (KZM), the universality of spontaneous defect generation in continuous phase transitions (CPTs) can be understood by the critical slowing down. In most CPTs of atomic Bose-Einstein condensates (BECs), the universality of spontaneous defect generations has been explained by the divergent relaxation time associated with the nontrivial gapless Bogoliubov excitations. However, for atomic BECs in synthetic gauge fields, their spontaneous superfluidity breakdown is resulted from the divergent correlation length associated with the zero Landau critical velocity. Here, by considering an atomic BEC ladder subjected to a synthetic magnetic field, we reveal that the spontaneous superfluidity breakdown obeys the KZM. The Kibble-Zurek scalings are derived from the Landau critical velocity which determines the correlation length. In further, the critical exponents are numerically extracted from the critical spatial-temporal dynamics of the bifurcation delay and the spontaneous vortex generation. Our study provides a general way to explore and understand the spontaneous superfluidity breakdown in CPTs from a single-well dispersion to a double-well one, such as, BECs in synthetic gauge fields, spin-orbit coupled BECs, and BECs in shaken optical lattices.

cond-mat.quant-gas↗

arXiv2016

Isospin properties of quark matter from a 3-flavor NJL model

We have studied the properties of hot and dense quark matter based on the 3-flavor Nambu-Jona-Lasinio (NJL) model as well as its Polyakov-loop extension (pNJL) with scalar-isovector and vector-isovector couplings. Provided a considerable large isospin asymmetry or isospin chemical potential, isospin splittings of constituent mass, chiral phase transition boundary, and critical point for $u$ and $d$ quarks can be observed for positive isovector coupling constants but are suppressed for negative ones. The quark matter symmetry energy decreases with the increasing isovector coupling constant, and is mostly enhanced in the pNJL model than in the NJL model. A positive scalar-isovector coupling constant is more likely to lead to an unstable isospin asymmetric quark matter. The isovector coupling has been further found to affect particle fractions as well as the equation of state in hybrid stars. Possible effects on the isospin properties of quark matter have also been discussed if the strangeness sector is further broken among the flavor symmetry.

nucl-th↗

arXiv2016

A Study of MatchPyramid Models on Ad-hoc Retrieval

Deep neural networks have been successfully applied to many text matching tasks, such as paraphrase identification, question answering, and machine translation. Although ad-hoc retrieval can also be formalized as a text matching task, few deep models have been tested on it. In this paper, we study a state-of-the-art deep matching model, namely MatchPyramid, on the ad-hoc retrieval task. The MatchPyramid model employs a convolutional neural network over the interactions between query and document to produce the matching score. We conducted extensive experiments to study the impact of different pooling sizes, interaction functions and kernel sizes on the retrieval performance. Finally, we show that the MatchPyramid models can significantly outperform several recently introduced deep matching models on the retrieval task, but still cannot compete with the traditional retrieval models, such as BM25 and language models.

cs.IR↗

arXiv2016

Spin transport in intermediate-energy heavy-ion collisions as a probe of in-medium spin-orbit interactions

The spin up-down splitting of collective flows in intermediate-energy heavy-ion collisions as a result of the nuclear spin-orbit interaction is investigated within a spin- and isospin-dependent Boltzmann-Uehing-Uhlenbeck transport model SIBUU12. Using a Skyrme-type spin-orbit coupling quadratic in momentum, we found that the spin splittings of the directed flow and elliptic flow are largest in peripheral Au+Au collisions at beam energies of about 100-200 MeV/nucleon, and the effect is considerable even in smaller systems especially for nucleons with high transverse momenta. The collective flows of light clusters of different spin states are also investigated using an improved dynamical coalescence model with spin. Our study can be important in understanding the properties of in-medium nuclear spin-orbit interactions once the spin-dependent observables proposed in this work can be measured.

nucl-th↗

arXiv2016

Equations of motion of test particles for solving the spin-dependent Boltzmann-Vlasov equation

A consistent derivation of the equations of motion (EOMs) of test particles for solving the spin-dependent Boltzmann-Vlasov equation is presented. The resulting EOMs in phase space are similar to the canonical equations in Hamiltonian dynamics, and the EOM of spin is the same as that in the Heisenburg picture of quantum mechanics. Considering further the quantum nature of spin and choosing the direction of total angular momentum in heavy-ion reactions as a reference of measuring nucleon spin, the EOMs of spin-up and spin-down nucleons are given separately. The key elements affecting the spin dynamics in heavy-ion collisions are identified. The resulting EOMs provide a solid foundation for using the test-particle approach in studying spin dynamics in heavy-ion collisions at intermediate energies. Future comparisons of model simulations with experimental data will help constrain the poorly known in-medium nucleon spin-orbit coupling relevant for understanding properties of rare isotopes and their astrophysical impacts.

nucl-th↗

arXiv2016

Incorporating Semantic Knowledge into Latent Matching Model in Search

The relevance between a query and a document in search can be represented as matching degree between the two objects. Latent space models have been proven to be effective for the task, which are often trained with click-through data. One technical challenge with the approach is that it is hard to train a model for tail queries and tail documents for which there are not enough clicks. In this paper, we propose to address the challenge by learning a latent matching model, using not only click-through data but also semantic knowledge. The semantic knowledge can be categories of queries and documents as well as synonyms of words, manually or automatically created. Specifically, we incorporate semantic knowledge into the objective function by including regularization terms. We develop two methods to solve the learning task on the basis of coordinate descent and gradient descent respectively, which can be employed in different settings. Experimental results on two datasets from an app search engine demonstrate that our model can make effective use of semantic knowledge, and thus can significantly enhance the accuracies of latent matching models, particularly for tail queries.

cs.IR↗

arXiv2016

Match-SRNN: Modeling the Recursive Matching Structure with Spatial RNN

Semantic matching, which aims to determine the matching degree between two texts, is a fundamental problem for many NLP applications. Recently, deep learning approach has been applied to this problem and significant improvements have been achieved. In this paper, we propose to view the generation of the global interaction between two texts as a recursive process: i.e. the interaction of two texts at each position is a composition of the interactions between their prefixes as well as the word level interaction at the current position. Based on this idea, we propose a novel deep architecture, namely Match-SRNN, to model the recursive matching structure. Firstly, a tensor is constructed to capture the word level interactions. Then a spatial RNN is applied to integrate the local interactions recursively, with importance determined by four types of gates. Finally, the matching score is calculated based on the global interaction. We show that, after degenerated to the exact matching scenario, Match-SRNN can approximate the dynamic programming process of longest common subsequence. Thus, there exists a clear interpretation for Match-SRNN. Our experiments on two semantic matching tasks showed the effectiveness of Match-SRNN, and its ability of visualizing the learned matching structure.

cs.CL↗