arXiv subjects

Tao Xiang

Publications and source records attributed to Tao Xiang.

At least 343 records · Page 19Linked to original sources

arXiv2017

Quantum phase transition in a multiconnected Jaynes-Cummings lattice

The rapid progress in quantum technology enables the implementation of artificial many-body systems with correlated photons and polaritons. A multiconnected Jaynes-Cummings (MCJC) lattice can be constructed by connecting qubits and cavities alternatively. Such kind of models can be realized with superconducting qubits coupled to superconducting microwave resonators or with quantum dots coupled to optical nanocavities. We study physical properties of the one-dimensional MCJC lattice using the density-matrix renormalization group method. This model has an intrinsic symmetry between the left and right qubit-cavity couplings. The competition between these couplings may drive the ground state either to a Mott-insulating or to a superfluid phase at an integer filling. We calculate the single-particle and density-density correlation functions, the correlation lengths in the Mott-insulating phase and the Luttinger parameters in the superfluid phase, and determine accurately the critical points that separate these two phases.

quant-ph↗

arXiv2017

Anisotropic electron-phonon coupling in the spinel oxide superconductor

Among hundreds of spinel oxides, LiTi2O4 (LTO) is the only one that exhibits superconductivity (Tc ~13 K). Although the general electron-phonon coupling is still the main mechanism for electron pairing in LTO, unconventional behaviors such as the anomalous magnetoresistance, anisotropic orbital/spin susceptibilities, etc. reveal that both the spin and the orbital interactions should also be considered for understanding the superconductivity. Here, we investigate tunneling spectra of [111]-, [110]- and [001]-oriented high quality LTO thin films. Several bosonic modes in tunneling spectra are observed in the [111]- and [110]-oriented films but not in [001]-oriented ones, and these modes still exist at T = 2Tc and beyond the upper critical field, which are confirmed as stemming from electron-phonon interaction by DFT calculations. These modes only appear in special surface orientations, indicating that the electron-phonon coupling in LTO system is highly anisotropic and may be enhanced by orbital-related state. The anisotropic electron-phonon coupling should be taken seriously in understanding the nature of LTO superconductivity.

cond-mat.supr-con↗

arXiv2016

Highly Efficient Regression for Scalable Person Re-Identification

Existing person re-identification models are poor for scaling up to large data required in real-world applications due to: (1) Complexity: They employ complex models for optimal performance resulting in high computational cost for training at a large scale; (2) Inadaptability: Once trained, they are unsuitable for incremental update to incorporate any new data available. This work proposes a truly scalable solution to re-id by addressing both problems. Specifically, a Highly Efficient Regression (HER) model is formulated by embedding the Fisher's criterion to a ridge regression model for very fast re-id model learning with scalable memory/storage usage. Importantly, this new HER model supports faster than real-time incremental model updates therefore making real-time active learning feasible in re-id with human-in-the-loop. Extensive experiments show that such a simple and fast model not only outperforms notably the state-of-the-art re-id methods, but also is more scalable to large data with additional benefits to active learning for reducing human labelling effort in re-id deployment.

cs.CV↗

arXiv2016

A generalized Lanczos method for systematic optimization of tensor network states

We propose a generalized Lanczos method to generate the many-body basis states of quantum lattice models using tensor-network states (TNS). The ground-state wave function is represented as a linear superposition composed from a set of TNS generated by Lanczos iteration. This method improves significantly both the accuracy and the efficiency of the tensor-network algorithm and allows the ground state to be determined accurately using TNS with very small virtual bond dimensions. This state contains significantly more entanglement than each individual TNS, reproducing correctly the logarithmic size dependence of the entanglement entropy in a critical system. The method can be generalized to non-Hamiltonian systems and to the calculation of low-lying excited states, dynamical correlation functions, and other physical properties of strongly correlated systems.

cond-mat.str-el↗

arXiv2016

Deep Transfer Learning for Person Re-identification

Person re-identification (Re-ID) poses a unique challenge to deep learning: how to learn a deep model with millions of parameters on a small training set of few or no labels. In this paper, a number of deep transfer learning models are proposed to address the data sparsity problem. First, a deep network architecture is designed which differs from existing deep Re-ID models in that (a) it is more suitable for transferring representations learned from large image classification datasets, and (b) classification loss and verification loss are combined, each of which adopts a different dropout strategy. Second, a two-stepped fine-tuning strategy is developed to transfer knowledge from auxiliary datasets. Third, given an unlabelled Re-ID dataset, a novel unsupervised deep transfer learning model is developed based on co-training. The proposed models outperform the state-of-the-art deep Re-ID models by large margins: we achieve Rank-1 accuracy of 85.4\%, 83.7\% and 56.3\% on CUHK03, Market1501, and VIPeR respectively, whilst on VIPeR, our unsupervised model (45.1\%) beats most supervised models.

cs.CV↗

arXiv2016

Semantic Regularisation for Recurrent Image Annotation

The "CNN-RNN" design pattern is increasingly widely applied in a variety of image annotation tasks including multi-label classification and captioning. Existing models use the weakly semantic CNN hidden layer or its transform as the image embedding that provides the interface between the CNN and RNN. This leaves the RNN overstretched with two jobs: predicting the visual concepts and modelling their correlations for generating structured annotation output. Importantly this makes the end-to-end training of the CNN and RNN slow and ineffective due to the difficulty of back propagating gradients through the RNN to train the CNN. We propose a simple modification to the design pattern that makes learning more effective and efficient. Specifically, we propose to use a semantically regularised embedding layer as the interface between the CNN and RNN. Regularising the interface can partially or completely decouple the learning problems, allowing each to be more effectively trained and jointly training much more efficient. Extensive experiments show that state-of-the art performance is achieved on multi-label classification as well as image captioning.

cs.CV↗

arXiv2016

The vital role of hole-carriers for superconductivity in pressurized black phosphorus

The influence of carrier type on superconductivity has been an important issue for understanding both conventional and unconventional superconductors [1-7]. For elements that superconduct, it is known that hole-carriers govern the superconductivity for transition and main group metals [8-10]. The role of hole-carriers in elements that are not normally conducting but can be converted to superconductors, however, remains unclear due to the lack of experimental data. Here we report the first in-situ high pressure Hall effect measurements on single crystal black phosphorus, measured up to ~ 50 GPa, and find a correlation between the Hall coefficient and the superconducting transition temperature (TC). Our results reveal that hole-carriers play a vital role in developing superconductivity and enhancing TC. Importantly, we also find a Lifshitz transition in the high-pressure cubic phase at ~17.2GPa, which uncovers the origin of a puzzling valley in the superconducting TC-pressure phase diagram. These results offer insight into the role of hole-carriers in developing superconductivity in simple semiconducting solids under pressure.

cond-mat.supr-con↗

arXiv2016

A conducting nano-filament (CNF) network as a precursor to the origin of superconductivity in electron-doped copper oxides

Emergency of superconductivity at the instabilities of antiferromagnetism has been widely recognized in unconventional superconductors. In copper-oxide superconductors, spin fluctuations play a predominant role in electron pairing with electron dopants yet composite orders veil the nature of superconductivity for hole-doped family. However, in electron-doped copper oxide superconductors (cuprates) the AFM critical end point is still in controversy for different probes, demonstrating high sensitivity to oxygen content. Here, by carefully tuning the oxygen content, a systematic study of the Hall signal and magnetoresistivity up to 58 Tesla on LCCO thin films identifies two characteristic temperatures. The former is quite robust, whereas the latter becomes flexible with increasing magnetic field, thereby linking respectively to two- and three-dimensional AFM, evident from the multidimensional phase diagram as a function of oxygen and Ce dopants. A rigorous theoretical analysis of the presented data suggest the existence of conductive nano-filamentary structures that effectively corroborate all previously reported field studies. The new findings provide a uniquely consistent alternative picture in understanding the interactions between AFM and superconductivity in electron-doped cuprates and offer a consolidating interpretation to the pioneering scaling law in cuprates recently established by Bozovic et al. (Nature, 2016)

cond-mat.mes-hall↗

arXiv2016

Self-consistent spin-wave analysis of the 1/3 magnetization plateau in the kagome antiferromagnet

We propose a modified spin-wave theory to study the 1/3 magnetization plateau of the antiferromagnetic Heisenberg model on the kagome lattice. By the self-consistent inclusion of quantum corrections, the 1/3 plateau is stabilized over a broad range of magnetic fields for all spin quantum numbers, S. The values of the critical magnetic fields and the widths of the magnetization plateaus are fully consistent with recent numerical results from exact diagonalization and infinite projected entangled paired states.

cond-mat.str-el↗

arXiv2016

Learning a Discriminative Null Space for Person Re-identification

Most existing person re-identification (re-id) methods focus on learning the optimal distance metrics across camera views. Typically a person's appearance is represented using features of thousands of dimensions, whilst only hundreds of training samples are available due to the difficulties in collecting matched training images. With the number of training samples much smaller than the feature dimension, the existing methods thus face the classic small sample size (SSS) problem and have to resort to dimensionality reduction techniques and/or matrix regularisation, which lead to loss of discriminative power. In this work, we propose to overcome the SSS problem in re-id distance metric learning by matching people in a discriminative null space of the training data. In this null space, images of the same person are collapsed into a single point thus minimising the within-class scatter to the extreme and maximising the relative between-class separation simultaneously. Importantly, it has a fixed dimension, a closed-form solution and is very efficient to compute. Extensive experiments carried out on five person re-identification benchmarks including VIPeR, PRID2011, CUHK01, CUHK03 and Market1501 show that such a simple approach beats the state-of-the-art alternatives, often by a big margin.

cs.CV↗

arXiv2015

First-principles study of FeSe epitaxial films on SrTiO3

The discovery of high temperature superconductivity in FeSe films on SrTiO3 substrate has inspired great experimental and theoretical interests. First-principles density functional theory calculations, which have played an important role in the study of bulk iron-based superconductors, also participate in the investigation of interfacial superconductivity. In this article, we review the calculation results on the electronic and magnetic structures of FeSe epitaxial films, emphasizing on the interplay between different degrees of freedom, such as charge, spin, and lattice vibrations. Furthermore, the comparison between FeSe monolayer and bilayer films on SrTiO3 is discussed.

cond-mat.supr-con↗

arXiv2015

A close look at antiferromagnetism in multidimensional phase diagram of electron-doped copper oxide

Emergency of superconductivity at the instabilities of antiferromagnetism (AFM), spin/charge density waves has been widely recognized in unconventional superconductors. In copper-oxide superconductors, spin fluctuations play a predominant role in electron pairing with electron dopants yet composite orders veil the nature of superconductivity for hole-doped family. However, in electron-doped ones the ending point of AFM is still in controversy for different probes or its sensitivity to oxygen content. Here, by carefully tuning the oxygen content, a systematic study of Hall signal and magnetoresistivity up to 58 Tesla on optimally doped La2-xCexCuO4+-δ (x = 0.10) thin films identifies two characteristic temperatures at 62.5+-7.5 K and 25+-5 K. The former is quite robust whereas the latter becomes flexible with increasing magnetic field, thereby linked to two- and three-dimensional AFM, evident from the multidimensional phase diagram as a function of oxygen as well as Ce dopants. Consequently, the observation of extended AFM phase in contrast to μSR probe corroborates an elevated critical doping in field, providing an unambiguous picture to understand the interactions between AFM and superconductivity.

cond-mat.supr-con↗

arXiv2015

Nematic antiferromagnetic states in bulk FeSe

We revisit bulk FeSe through the systematic first-principles electronic structure calculations. We find that there are a series of staggered $n$-mer antiferromagnetic (AFM) states with corresponding energies below that of the collinear AFM state which is the ground state for the parent compounds of most iron-based superconductors. Here the staggered $n$-mer ($n$ any integer $>1$) means that a set of $n$ adjacent spins parallel on a line along $b$-axis with spins in antiparallel between $n$-mers and along $a$-axis. Among them, the lowest energy states are quasi-degenerate staggered dimer and staggered trimer AFM states as well as their any staggered combinations. Thus, to have the largest entropy to minimize the free energy at low temperature, the most favorable state is such a quasi-one-dimensional antiferromagnet in which along $b$-axis a variety of $n$-mers, mostly dimers and trimers, are randomly antiparallel aligned while along $a$-axis spins are antiparallel aligned, i.e. actually a nematic paramagnet. This finding accounts well for the absence of long-range magnetic order in bulk FeSe and meanwhile indicates the dominant stripe spin fluctuation and the nematicity as spin-driven.

cond-mat.supr-con↗

arXiv2015

Ground State Degeneracy of Interacting Spinless Fermions

We propose an eigen-operator scheme to study the lattice model of interacting spinless fermions at half filling and show that this model possesses a hidden form of reflection positivity in its Majorana fermion representation. Based on this observation, we prove rigourously that the ground state of this model is either unique or doubly degenerate if the lattice size $N$ is even, and is always doubly degenerate if $N$ is odd. This proof holds in all dimensions with arbitrary lattice structures.

cond-mat.str-el↗

arXiv2015

Tensor network algorithm by coarse-graining tensor renormalization on finite periodic lattices

We develop coarse-graining tensor renormalization group algorithms to compute physical properties of two-dimensional lattice models on finite periodic lattices. Two different coarse-graining strategies, one based on the tensor renormalization group and the other based on the higher-order tensor renormalization group, are introduced. In order to optimize the tensor-network model globally, a sweeping scheme is proposed to account for the renormalization effect from the environment tensors under the framework of second renormalization group. We demonstrate the algorithms by the classical Ising model on the square lattice and the Kitaev model on the honeycomb lattice, and show that the finite-size algorithms achieve substantially more accurate results than the corresponding infinite-size ones.

cond-mat.str-el↗

arXiv2015

Robust Subjective Visual Property Prediction from Crowdsourced Pairwise Labels

The problem of estimating subjective visual properties from image and video has attracted increasing interest. A subjective visual property is useful either on its own (e.g. image and video interestingness) or as an intermediate representation for visual recognition (e.g. a relative attribute). Due to its ambiguous nature, annotating the value of a subjective visual property for learning a prediction model is challenging. To make the annotation more reliable, recent studies employ crowdsourcing tools to collect pairwise comparison labels because human annotators are much better at ranking two images/videos (e.g. which one is more interesting) than giving an absolute value to each of them separately. However, using crowdsourced data also introduces outliers. Existing methods rely on majority voting to prune the annotation outliers/errors. They thus require large amount of pairwise labels to be collected. More importantly as a local outlier detection method, majority voting is ineffective in identifying outliers that can cause global ranking inconsistencies. In this paper, we propose a more principled way to identify annotation outliers by formulating the subjective visual property prediction task as a unified robust learning to rank problem, tackling both the outlier detection and learning to rank jointly. Differing from existing methods, the proposed method integrates local pairwise comparison labels together to minimise a cost that corresponds to global inconsistency of ranking order. This not only leads to better detection of annotation outliers but also enables learning with extremely sparse annotations. Extensive experiments on various benchmark datasets demonstrate that our new approach significantly outperforms state-of-the-arts alternatives.

cs.CV↗

arXiv2015

Sketch-a-Net that Beats Humans

We propose a multi-scale multi-channel deep neural network framework that, for the first time, yields sketch recognition performance surpassing that of humans. Our superior performance is a result of explicitly embedding the unique characteristics of sketches in our model: (i) a network architecture designed for sketch rather than natural photo statistics, (ii) a multi-channel generalisation that encodes sequential ordering in the sketching process, and (iii) a multi-scale network ensemble with joint Bayesian fusion that accounts for the different levels of abstraction exhibited in free-hand sketches. We show that state-of-the-art deep networks specifically engineered for photos of natural objects fail to perform well on sketch recognition, regardless whether they are trained using photo or sketch. Our network on the other hand not only delivers the best performance on the largest human sketch dataset to date, but also is small in size making efficient training possible using just CPUs.

cs.CV↗

arXiv2015

Evolution of electronic states in n-type copper oxide superconductor via electric double layer gating

Since the discovery of n-type copper oxide superconductors, the evolution of electron- and hole-bands and its relation to the superconductivity have been seen as a key factor in unveiling the mechanism of high-Tc superconductors. So far, the occurrence of electrons and holes in n-type copper oxides has been achieved by chemical doping, pressure, and/or deoxygenation. However, the observed electronic properties are blurred by the concomitant effects such as change of lattice structure, disorder, etc. Here, we report on successful tuning the electronic band structure of n-type Pr2-xCexCuO4 (x = 0.15) ultrathin films, via the electric double layer transistor technique. Abnormal transport properties, such as multiple sign reversals of Hall resistivity in normal and mixed states, have been revealed within an electrostatic field in range of -2 V to +2 V, as well as varying the temperature and magnetic field. In the mixed state, the intrinsic anomalous Hall conductivity invokes the contribution of both electron and hole-bands as well as the energy dependent density of states near the Fermi level. The two-band model can also describe the normal state transport properties well, whereas the carrier concentrations of electrons and holes are always enhanced or depressed simultaneously in electric fields. This is in contrast to the scenario of Fermi surface reconstruction by antiferromagnetism, where an anti-correlation between electrons and holes is commonly expected. Our findings paint the picture where Coulomb repulsion plays an important role in the evolution of the electronic states in n-type cuprate superconductors.

cond-mat.supr-con↗