Searcharxiv⌕ Search

arXiv subjects

Lei Shu

Publications and source records attributed to Lei Shu.

At least 91 records · Page 5Linked to original sources
arXiv2017

DOC: Deep Open Classification of Text Documents

Traditional supervised learning makes the closed-world assumption that the classes appeared in the test data must have appeared in training. This also applies to text learning or text classification. As learning is used increasingly in dynamic open environments where some new/test documents may not belong to any of the training classes, identifying these novel documents during classification presents an important problem. This problem is called open-world classification or open classification. This paper proposes a novel deep learning based approach. It outperforms existing state-of-the-art techniques dramatically.

cs.CL↗
arXiv2017

Discovery of slow magnetic fluctuations and critical slowing down in the pseudogap phase of YBa$_2$Cu$_3$O$_y$

Evidence for intra-unit-cell (IUC) magnetic order in the pseudogap region of high-$T_c$ cuprates below a temperature $T^\ast$ is found in several studies, but NMR and $μ$SR experiments do not observe the expected static local magnetic fields. It has been noted, however, that such fields could be averaged by fluctuations. Our measurements of muon spin relaxation rates in single crystals of YBa$_2$Cu$_3$O$_y$ reveal magnetic fluctuations of the expected order of magnitude that exhibit critical slowing down at $T^\ast$. These results are strong evidence for fluctuating IUC magnetic order in the pseudogap phase.

cond-mat.supr-con↗
arXiv2017

Supervised Complementary Entity Recognition with Augmented Key-value Pairs of Knowledge

Extracting opinion targets is an important task in sentiment analysis on product reviews and complementary entities (products) are one important type of opinion targets that may work together with the reviewed product. In this paper, we address the problem of Complementary Entity Recognition (CER) as a supervised sequence labeling with the capability of expanding domain knowledge as key-value pairs from unlabeled reviews, by automatically learning and enhancing knowledge-based features. We use Conditional Random Field (CRF) as the base learner and augment CRF with knowledge-based features (called the Knowledge-based CRF or KCRF for short). We conduct experiments to show that KCRF effectively improves the performance of supervised CER task.

cs.CL↗
arXiv2017

Lifelong Learning CRF for Supervised Aspect Extraction

This paper makes a focused contribution to supervised aspect extraction. It shows that if the system has performed aspect extraction from many past domains and retained their results as knowledge, Conditional Random Fields (CRF) can leverage this knowledge in a lifelong learning manner to extract in a new domain markedly better than the traditional CRF without using this prior knowledge. The key innovation is that even after CRF training, the model can still improve its extraction with experiences in its applications.

cs.CL↗
arXiv2017

Investigation of the complex magnetism in the breathing pyrochlore LiIn(Cr$_{1-x}$Rh$_x$)$_4$O$_8$

We have performed a detailed investigation of the new `breathing' pyrochlore compound LiInCr$_4$O$_8$ through Rh substitution with measurements of magnetic susceptibility, specific heat, and x-ray powder diffraction. The antiferromagnetic phase of LiInCr$_4$O$_8$ is found to be slowly suppressed with increasing Rh, up to the critical concentration of $x$ = 0.1 where the antiferromagnetic phase is still observed with the peak in specific heat $T_p$ = 12.5 K, slightly lower than $T_p$ = 14.3 K for the $x$ = 0 compound. From the measurements of magnetization we also uncover evidence that substitution increases the amount of frustration. Comparisons are made with the LiGa$_y$In$_{1-y}$Cr$_4$O$_8$ system as well as other frustrated pyrochlore-related materials and find comparable amounts of frustration. The results of this work shows that engineered breathing pyrochlores presents an important method to further understand the complex magnetism in frustrated systems.

cond-mat.str-el↗
arXiv2016

Supervised Opinion Aspect Extraction by Exploiting Past Extraction Results

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. For example, "screen" is an aspect in iPhone, but not only iPhone has a screen, many electronic devices have screens too. When "screen" appears in a review of a new domain (or product), it is likely to be an aspect too. Knowing this information enables us to do much better extraction in the new domain. This paper proposes a novel extraction method exploiting this idea in the context of supervised sequence labeling. Experimental results show that it produces markedly better results than without using the past information.

cs.CL↗
arXiv2016

Mining Compatible/Incompatible Entities from Question and Answering via Yes/No Answer Classification using Distant Label Expansion

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., via a yes/no question "Does this mouse work with windows 10?". In this paper, we address the problem of extracting compatible and incompatible products from yes/no questions in PCQA. This problem can naturally have a two-stage framework: first, we perform Complementary Entity (product) Recognition (CER) on yes/no questions; second, we identify the polarities of yes/no answers to assign the complementary entities a compatibility label (compatible, incompatible or unknown). We leverage an existing unsupervised method for the first stage and a 3-class classifier by combining a distant PU-learning method (learning from positive and unlabeled examples) together with a binary classifier for the second stage. The benefit of using distant PU-learning is that it can help to expand more implicit yes/no answers without using any human annotated data. We conduct experiments on 4 products to show that the proposed method is effective.

cs.CL↗
arXiv2016

CER: Complementary Entity Recognition via Knowledge Expansion on Large Unlabeled Product Reviews

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. In this paper, we address the problem of Complementary Entity Recognition (CER). Since no existing method can solve this problem, we first propose a novel unsupervised method to utilize syntactic dependency paths to recognize complementary entities. Then we expand category-level domain knowledge about complementary entities using only a few general seed verbs on a large amount of unlabeled reviews. The domain knowledge helps the unsupervised method to adapt to different products and greatly improves the precision of the CER task. The advantage of the proposed method is that it does not require any labeled data for training. We conducted experiments on 7 popular products with about 1200 reviews in total to demonstrate that the proposed approach is effective.

cs.CL↗
arXiv2016

Superconducting gap structure in ambient-pressure-grown $\mathrm{LaO_{0.5}F_{0.5}BiS_{2}}$

We have performed transverse-field muon spin relaxation (TF-$μ$SR) measurements on ambient-pressure-grown polycrystalline $\mathrm{LaO_{0.5}F_{0.5}BiS_{2}}$. From these measurements, no signature of magnetic order is found down to 25 mK. The value of the magnetic penetration depth extrapolated to 0 K is 0.89 (5) $μ$m. The temperature dependence of superconducting penetration depth is best described by either a multigap s + s-wave model with $Δ_{1}$ = 0.947 (7) meV and $Δ_{2}$ = 0.22 (4) meV or the ansiotropic s-wave model with $Δ(0)$ = 0.776 meV and anisotropic gap amplitude ratio $Δ_{min}/Δ_{max}$ = 0.34. Comparisons with other potentially multigap $\mathrm{BiS_{2}}$-based superconductors are discussed. We find that these $\mathrm{BiS_{2}}$-based superconductors, including $\mathrm{Bi_{4}O_{4}S_3}$ and the high-pressure synthesized $\mathrm{LaO_{0.5}F_{0.5}BiS_{2}}$, generally conform to the Uemura relation.

cond-mat.supr-con↗
arXiv2016

Quantum Criticality and Inhomogeneous Magnetic Order in Fe-doped alpha-YbAlB4

The intermediate-valent polymorphs $α$- and $β$-YbAlB$_4$ exhibit quantum criticality and other novel properties not usually associated with intermediate valence. Iron doping induces quantum criticality in $α$-YbAlB$_4$ and magnetic order in both compounds. We report results of muon spin relaxation ($μ$SR) experiments in the intermediate-valent alloys $α$-YbAl$_{1-x}$Fe$_x$B$_4$, $x = 0.014$ and 0.25. For $x = 0.014$ we find no evidence for magnetic order down to 25 mK\@. The dynamic muon spin relaxation rate $λ_d$ exhibits a power-law temperature dependence $λ_d \propto T^{-a}$, $a = 0.40(4)$, in the temperature range 100 mK--2 K, in disagreement with predictions by theories of antiferromagnetic (AFM) or valence quantum critical behavior. For $x = 0.25$, where AFM order develops in the temperature range 7.5--10 K, where we find coexistence of meso- or macroscopically segregated paramagnetic and AFM phases, with considerable disorder in the latter down to 2 K.

cond-mat.str-el↗
arXiv2016

Optimal Design of Compact Receive Array in Industrial Wireless Sensor Networks

With the development of wireless communication, industrial wireless sensor networks (IWSNs) plays an important role in monitoring and control systems. In this paper, we extend the application of IWSNs into High Frequency Surface-Wave Radar (HFSWR) system. The traditional antenna is replaced by mobile IWSNs. In combination of the application precondition of super-directivity in HF band and circular topology of IWSNs, a super-directivity synthesis method is presented for designing super-directivity array. In this method, the dominance of external noise is ensured by constraining the Ratio of External to Internal Noise (REIN) of the array, and the desired side lobe level is achieved by implementing linear constraint. By using this method, the highest directivity will be achieved in certain conditions. Using the designed super directive circular array as sub-arrays, the compact receive antenna array is constructed, the purpose of miniaturization is achieved. Simulation verifies that the proposed method is correct and effective, the validity of the proposed method has been proved.

cs.IT↗
arXiv2015

Unstable Spin-Ice Order in the Stuffed Metallic Pyrochlore Pr$_{2+x}$Ir$_{2-x}$O$_{7-δ}$

Specific heat, elastic neutron scattering, and muon spin rotation ($μ$SR) experiments have been carried out on a well-characterized sample of "stuffed" (Pr-rich) Pr$_{2+x}$Ir$_{2-x}$O$_{7-δ}$. Elastic neutron scattering shows the onset of long-range spin-ice "2-in/2-out" magnetic order at $T_M = 0.93$ K, with an ordered moment of 1.7(1)$μ_\mathrm{B}$/Pr ion at low temperatures. Approximate lower bounds on the correlation length and correlation time in the ordered state are 170 Å and 0.7 ns, respectively. $μ$SR experiments yield an upper bound 2.6(7) mT on the local field $B_\mathrm{loc}^{4f}$ at the muon site, which is nearly two orders of magnitude smaller than the expected dipolar field for long-range spin-ice ordering of 1.7$μ_B$ moments (120--270 mT, depending on muon site). This shortfall is due in part to splitting of the non-Kramers crystal-field ground-state doublets of near-neighbor Pr$^{3+}$ ions by the $μ^+$-induced lattice distortion. For this to be the only effect, however, $\sim$160 Pr moments out to a distance of $\sim$14 Å must be suppressed. An alternative scenario, which is consistent with the observed reduced nuclear hyperfine Schottky anomaly in the specific heat, invokes slow correlated Pr-moment fluctuations in the ordered state that average $B_\mathrm{loc}^{4f}$ on the $μ$SR time scale (${\sim}10^{-7}$ s), but are static on the time scale of the elastic neutron scattering experiments (${\sim}10^{-9}$ s). In this picture the dynamic muon relaxation suggests a Pr$^{3+}$ $4f$ correlation time of a few nanoseconds, which should be observable in a neutron spin echo experiment.

cond-mat.str-el↗
arXiv2015

Broken Time Reversal Symmetry in Superconducting Pr1-xCexPt4Ge12

We report results of zero-field muon spin relaxation experiments on the filled-skutterudite superconductors~Pr$_{1-x}$Ce$_{x}$Pt$_4$Ge$_{12}$, $x = 0$, 0.07, 0.1, and 0.2, to investigate the effect of Ce doping on broken time-reversal symmetry (TRS) in the superconducting state. In these alloys broken TRS is signaled by the onset of a spontaneous static local magnetic field~$B_s$ below the superconducting transition temperature. We find that $B_s$ decreases linearly with $x$ and $\to 0$ at $x \approx 0.4$, close to the concentration above which superconductivity is no longer observed. The (Pr,Ce)Pt$_4$Ge$_{12}$ and isostructural (Pr,La)Os$_4$Sb$_{12}$ alloy series both exhibit superconductivity with broken TRS, and in both the decrease of $B_s$ is proportional to the decrease of Pr concentration. This suggests that Pr-Pr intersite interactions are responsible for the broken TRS\@. The two alloy series differ in that the La-doped alloys are superconducting for all La concentrations, suggesting that in (Pr,Ce)Pt$_4$Ge$_{12}$ pair-breaking by Ce doping suppresses superconductivity. For all $x$ the dynamic muon spin relaxation rate decreases somewhat in the superconducting state. This may be due to Korringa relaxation by conduction electrons, which is reduced by the opening of the superconducting energy gap.

cond-mat.supr-con↗
arXiv2014

Landau Renormalizations of Superfluid density in Heavy Fermion Superconductor CeCoIn5

The formation of heavy fermion bands can occur by means of the conversion of a periodic array of local moments into itinerant electrons via the Kondo effect and the huge consequent Fermi-liquid renormalizations. Leggett predicted for liquid $^3$He that Fermi-liquid renormalizations change in the superconducting state, leading to a temperature dependence of the London penetration depth~$Λ$ quite different from that in the BCS theory. Using Leggett's theory, as modified for heavy fermions, it is possible to extract from the measured temperature dependence of $Λ$ in high quality samples both Landau parameters $F_0^s$ and $F_1^s$; this has never been accomplished before. A modification of the temperature dependence of the specific heat $C_\mathrm{el}$, related to that of $Λ$, is also expected. We have carefully determined the magnitude and temperature dependence of $Λ$ in CeCoIn$_5$ by muon spin relaxation rate measurements to obtain $F_0^s = 36 \pm 1$ and $F_1^s = 1.2 \pm 0.3$, and find a consistent change in the temperature dependence of electronic specific heat $C_\mathrm{el}$. This, the first determination of $F_1^s$ with a value~$\ll F_0^s$ in a heavy fermion compound, tests the basic assumption of the theory of heavy fermions, that the frequency dependence of the self-energy is much more important than its momentum dependence.

cond-mat.str-el↗
arXiv2014

Muon spin rotation and relaxation in Pr$_{1-x}$Nd$_x$Os$_4$Sb$_{12}$: Paramagnetic states

Positive-muon ($μ^+$) Knight shifts have been measured in the paramagnetic states of Pr$_{1-x}$Nd$_x$Os$_4$Sb$_{12}$ alloys, where $x =$ 0, 0.25, 0.45, 0.50, 0.55, 0.75, and 1.00. In Pr-substituted NdOs$_4$Sb$_{12}$ ($x \le$ 0.75), but not in NdOs$_4$Sb$_{12}$, Clogston-Jaccarino plots of $μ^+$ Knight shift~$K$ versus magnetic susceptibility~$χ$ exhibit an anomalous saturation of $K(χ)$ at $\sim-$0.5% for large susceptibilities (low temperatures), indicating a reduction of the coupling strength between $μ^+$ spins and $4f$ paramagnetism for temperatures $\lesssim$ 15~K. We speculate that itinerant Pr$^{3+}$ quadrupolar excitations, invoked to mediate the superconducting Cooper-pair interaction, might modify the $μ^+$-$4f$ ion indirect spin-spin interaction.

cond-mat.str-el↗
arXiv2014

Muon spin rotation and relaxation in Pr$_{1-x}$Nd$_x$Os$_4$Sb$_{12}$: Magnetic and superconducting ground states

Muon spin rotation and relaxation ($μ$SR) experiments have been carried out to characterize magnetic and superconducting ground states in the Pr$_{1-x}$Nd$_x$Os$_4$Sb$_{12}$ alloy series. In the ferromagnetic end compound NdOs$_4$Sb$_{12}$ the spontaneous local field at positive-muon ($μ^+$) sites below the ordering temperature $T_C$ is greater than expected from dipolar coupling to ferromagnetically aligned Nd$^{3+}$ moments, indicating an additional indirect RKKY-like transferred hyperfine mechanism. For 0.45 $\le x \le$ 0.75, $μ^+$ spin relaxation rates in zero and weak longitudinal applied fields indicate that static fields at $μ^+$ sites below $T_C$ are reduced and strongly disordered. We argue this is unlikely to be due to reduction of Nd$^{3+}$ moments, and speculate that the Nd$^{3+}$-$μ^+$ interaction is suppressed and disordered by Pr doping. In an $x$ = 0.25 sample, which is superconducting below $T_c$ = 1.3 K, there is no sign of "spin freezing" (static Nd$^{3+}$ magnetism), ordered or disordered, down to 25 mK. Dynamic $μ^+$ spin relaxation is strong, indicating significant Nd-moment fluctuations. The $μ^+$ diamagnetic frequency shift and spin relaxation in the superconducting vortex-lattice phase decrease slowly below $T_c$, suggesting pair breaking and/or possible modification of Fermi-liquid renormalization by Nd spin fluctuations. For 0.25 $\le x \le$ 0.75, the $μ$SR data provide evidence against phase separation; superconductivity and Nd$^{3+}$ magnetism coexist on the atomic scale.

cond-mat.str-el↗
arXiv2011

A Secured Health Care Application Architecture for Cyber-Physical Systems

Cyber-physical systems (CPS) can be viewed as a new generation of systems with integrated control, communication and computational capabilities. Like the internet transformed how humans interact with one another, cyber-physical systems will transform how people interact with the physical world. Currently, the study of CPS is still in its infancy and there exist many research issues and challenges ranging from electricity power, health care, transportation and smart building etc. In this paper, an introduction of CPeSC3 (cyber physical enhanced secured wireless sensor networks (WSNs) integrated cloud computing for u-life care) architecture and its application to the health care monitoring and decision support systems is given. The proposed CPeSC3 architecture is composed of three main components, namely 1) communication core, 2) computation core, and 3) resource scheduling and management core. Detailed analysis and explanation are given for relevant models such as cloud computing, real time scheduling and security models. Finally, a medical health care application scenario is presented based on our practical test-bed which has been built for 3 years.

cs.NI↗