arXiv subjects

Hui Liu

Publications and source records attributed to Hui Liu.

At least 523 records · Page 29Linked to original sources

arXiv2020

Spin polarization of electrons through corrugated surface in magnetic field

Noninteracting electrons confined to a corrugated surface are investigated in magnetic field, and the associated effective Pauli equation is given analytically by the thin-layer quantization scheme. Interestingly, the Zeeman splitting gaps can be adjusted by curvature, and there is a geometric potential induced by curvature. Further, we discuss the spin-dependent transport properties for confined electrons by numerical calculation. More interestingly, we find that the spin polarization induced by curvature becomes substantial when the incident energy has small value. The results are considerable for a spin transistor with small spin current.

cond-mat.mes-hall↗

arXiv2019

Exceptional Cones in 4D Parameter Space

The notion of synthetic dimensions has expanded the realm of topological physics to four dimensional (4D) space lately. In this work, non-Hermiticity is used as a synthetic parameter in PT-symmetric photonic crystals to study the topological physics in 4D non-Hermitian synthetic parameter space. We realize a 3D exceptional hypersurface (EHS) in such 4D parameter space, and the degeneracy points emerge due to the symmetry of synthetic parameters. We further demonstrate the existence of exceptional degenerate points (EDPs) on the EHS that originates from the chirality of exceptional points (EPs), and the exceptional surface near EDPs behaves like a Dirac cone. We further show that a very narrow reflection plateau can be found near these EDPs, and their sensitivity towards the PT-symmetry breaking environmental perturbation can make these degeneracy points useful in optical sensing and many other nonlinear and quantum optical applications.

physics.optics↗

arXiv2019

Clustering-aware Graph Construction: A Joint Learning Perspective

Graph-based clustering methods have demonstrated the effectiveness in various applications. Generally, existing graph-based clustering methods first construct a graph to represent the input data and then partition it to generate the clustering result. However, such a stepwise manner may make the constructed graph not fit the requirements for the subsequent decomposition, leading to compromised clustering accuracy. To this end, we propose a joint learning framework, which is able to learn the graph and the clustering result simultaneously, such that the resulting graph is tailored to the clustering task. The proposed model is formulated as a well-defined nonnegative and off-diagonal constrained optimization problem, which is further efficiently solved with convergence theoretically guaranteed. The advantage of the proposed model is demonstrated by comparing with 19 state-of-the-art clustering methods on 10 datasets with 4 clustering metrics.

cs.LG↗

arXiv2019

Realization of Photonic Charge-2 Dirac Point by Engineering Super-modes in Topological Superlattices

Quite recently a novel variety of unconventional fourfold linear band degeneracy points has been discovered in certain condensed-matter systems. Contrary to the standard 3-D Dirac monopoles, these quadruple points referred to as the charge-2 Dirac points are characterized by nonzero net topological charges, which can be exploited to delve into hitherto unknown realms of topological physics. Here, we report on the experimental realization of the charge-2 Dirac point by deliberately engineering hybrid topological states called super-modes in a 1-D optical superlattice system with two additional synthetic dimensions. Utilizing direct reflection and transmission measurements, we exhibit the existence of super-modes attributed to the synthetic charge-2 Dirac point, which has been achieved in the visible region for the first time. We also show the experimental approach to manipulating two spawned Weyl points that are identically charged in synthetic space. Moreover, topological end modes uniquely resulting from the charge-2 Dirac point can be delicately controlled within truncated superlattice samples, opening a pathway for us to rationally engineer local fields with intense enhancement.

physics.optics↗

arXiv2019

Pressure-induced superconductivity and structural transition in ferromagnetic Cr2Si2Te6

The discovery of intrinsic magnetism in atomically thin two-dimensional transition-metal trichalcogenides has attracted intense research interest due to the exotic properties of magnetism and potential applications in devices. Pressure has proven to be an effective tool to manipulate the crystal and electronic structures of the materials. Here, we report investigations on ferromagnetic van der Waals Cr2Si2Te6 via high-pressure synchrotron x-ray diffraction, electrical resistance, Hall resistance, and magnetoresistance measurements. Under compression, Cr2Si2Te6 simultaneously undergoes a structural transition, emergence of superconductivity at 3 K, sign change of the magnetoresistance, and dramatic change of the Hall coefficient at ~8 GPa. The superconductivity persists up to the highest measured pressure of 47.1 GPa with a maximum Tc = 4.5 K at ~30 GPa. The discovery of superconductivity in the two-dimensional van der Waals ferromagnetic Cr-based Cr2Si2Te6 provides new perspectives to explore superconductivity and the interplay between superconductivity and magnetism.

cond-mat.supr-con↗

arXiv2019

An Experimentally Verified Approach to non-Entanglement-Breaking Channel Certification

Ensuring the non-entanglement-breaking (non-EB) property of quantum channels is crucial for the effective distribution and storage of quantum states. However, a practical method for direct and accurate certification of the non-EB feature is highly desirable. Here, we propose and verify a realistic source based measurement device independent certification of non-EB channels. Our method is resilient to repercussions on the certification from experimental conditions, such as multiphotons and imperfect state preparation, and can be implemented with information incomplete set. We achieve good agreement between experimental outcomes and theoretical predictions, which is validated by the expected results of the ideal semi-quantum signaling game, and accurately certify the non-EB channels. Furthermore, our approach is highly robust to effects from noise. Therefore, the proposed approach can be expected to play a significant role in the design and evaluation of realistic quantum channels.

quant-ph↗

arXiv2019

Learning Multi-level Dependencies for Robust Word Recognition

Robust language processing systems are becoming increasingly important given the recent awareness of dangerous situations where brittle machine learning models can be easily broken with the presence of noises. In this paper, we introduce a robust word recognition framework that captures multi-level sequential dependencies in noised sentences. The proposed framework employs a sequence-to-sequence model over characters of each word, whose output is given to a word-level bi-directional recurrent neural network. We conduct extensive experiments to verify the effectiveness of the framework. The results show that the proposed framework outperforms state-of-the-art methods by a large margin and they also suggest that character-level dependencies can play an important role in word recognition.

cs.CL↗

arXiv2019

Stochastic Channel-Based Federated Learning for Medical Data Privacy Preserving

Artificial neural network has achieved unprecedented success in the medical domain. This success depends on the availability of massive and representative datasets. However, data collection is often prevented by privacy concerns and people want to take control over their sensitive information during both training and using processes. To address this problem, we propose a privacy-preserving method for the distributed system, Stochastic Channel-Based Federated Learning (SCBF), which enables the participants to train a high-performance model cooperatively without sharing their inputs. Specifically, we design, implement and evaluate a channel-based update algorithm for the central server in a distributed system, which selects the channels with regard to the most active features in a training loop and uploads them as learned information from local datasets. A pruning process is applied to the algorithm based on the validation set, which serves as a model accelerator. In the experiment, our model presents better performances and higher saturating speed than the Federated Averaging method which reveals all the parameters of local models to the server when updating. We also demonstrate that the saturating rate of performance could be promoted by introducing a pruning process. And further improvement could be achieved by tuning the pruning rate. Our experiment shows that 57% of the time is saved by the pruning process with only a reduction of 0.0047 in AUCROC performance and a reduction of 0.0068 in AUCPR.

cs.LG↗

arXiv2019

Elliptic and non-hyperbolic closed characteristics on compact convex P-cyclic symmetric hypersurfaces in ${\bf R}^{2n}$

Let $Σ$ be a compact convex hypersurface in ${\bf R}^{2n}$ which is P-cyclic symmetric, i.e., $x\in Σ$ implies $Px\inΣ$ with P being a $2n\times2n$ symplectic orthogonal matrix and $P^k=I_{2n}$, where $n, k\geq2$, $ker(P-I_{2n})=0$. In this paper, we first generalize Ekeland index theory for periodic solutions of convex Hamiltonian system to a index theory with P boundary value condition and study its relationship with Maslov P-index theory, then we use index theory to prove the existence of elliptic and non-hyperbolic closed characteristics on compact convex P-cyclic symmetric hypersurfaces in ${\bf R}^{2n}$ for a broad class of symplectic orthogonal matrix P.

math.DS↗

arXiv2019

Ground States of the $ϕ^4$ Double-Well QFT

At second order in perturbation theory, we find the ground states of the $ϕ^4$ double-well quantum field theory in 1+1 dimensions. The operators which create these ground states from the free vacuum are constructed explicitly at this order, as is the operator which interpolates between the ground states. As a warm up we perform the analogous calculation in quantum mechanics, where the true ground state is unique but in perturbation theory there are also two ground states.

hep-th↗

arXiv2019

Controlling Thermal Emission by Parity-symmetric Fano Resonance of Optical Absorbers in Metasurfaces

In this study, we designed a metasurface based on Al/SiN/Al nano-sandwich optical absorbers. Parity-symmetric dark magnetic resonance modes, anti-symmetric bright magnetic resonance modes, and surface lattice modes are established simultaneously in such a meta-surface. By changing the structural parameters, the wavelength of the dark mode can be tuned to overlap with the lattice mode. Fano resonance is introduced by coupling between the dark and lattice modes. Angular-resolved thermal emission experiments showed that the thermal emission induced by Fano resonance exhibited a narrow wavelength spectrum and directional radiation angle. Fano-type thermal emitters proposed in this work can possibly be used as coherent infrared light sources in the future.

physics.optics↗

arXiv2019

A Scalable Thermal Reservoir Simulator for Giant Models on Parallel Computers

This paper introduces the model, numerical methods, algorithms and parallel implementation of a thermal reservoir simulator that designed for numerical simulations of thermal reservoir with multiple components in three dimensional domain using distributed-memory parallel computers. Its full mathematical model is introduced with correlations for important properties and well modeling. Various well constraints, such as fixed bottom hole pressure, fixed oil, water, gas and liquid rates, constant heat transfer model, convective heat transfer model, heater model (temperature control, rate control, dual rate/temperature control), and subcool (steam trap), are introduced in details, including their mathematical models and methods. Efficient numerical methods and parallel computing technologies are presented. The simulator is designed for giant models with billions or even trillions of grid blocks using hundreds of thousands of CPUs. Numerical experiments show that our results match commercial simulators, which confirms the correctness of our methods and implementations. SAGD simulation with 15106 well pairs is also presented to study the effectiveness of our numerical methods. Scalability testings demonstrate that our simulator can handle giant models with over 200 billion grid blocks using 100,800 CPU cores and the simulator has good scalability.

cs.CE↗

arXiv2019

Adversarial Attacks and Defenses in Images, Graphs and Text: A Review

Deep neural networks (DNN) have achieved unprecedented success in numerous machine learning tasks in various domains. However, the existence of adversarial examples has raised concerns about applying deep learning to safety-critical applications. As a result, we have witnessed increasing interests in studying attack and defense mechanisms for DNN models on different data types, such as images, graphs and text. Thus, it is necessary to provide a systematic and comprehensive overview of the main threats of attacks and the success of corresponding countermeasures. In this survey, we review the state of the art algorithms for generating adversarial examples and the countermeasures against adversarial examples, for the three popular data types, i.e., images, graphs and text.

cs.LG↗

arXiv2019

Squeezing the Free Scalar Ground State

Consider two free Hamiltonians for the same scalar field with two different masses. Wefind a squeeze operator which maps the ground state of one to the other. The operatoris described in both the Dirac and also the Schrodinger wavefunctional formalismsfor quantum field theory. We conjecture that this construction can be generalized toobtain operators which map between distinct topological sectors in the same theory.

hep-th↗

arXiv2019

Privacy Preserving Stochastic Channel-Based Federated Learning with Neural Network Pruning

Artificial neural network has achieved unprecedented success in a wide variety of domains such as classifying, predicting and recognizing objects. This success depends on the availability of big data since the training process requires massive and representative data sets. However, data collection is often prevented by privacy concerns and people want to take control over their sensitive information during both training and using processes. To address this problem, we propose a privacy-preserving method for the distributed system, Stochastic Channel-Based Federated Learning (SCBF), which enables the participants to train a high-performance model cooperatively without sharing their inputs. We design, implement and evaluate a channel-based update algorithm for the central server in a distributed system, which selects the channels with regard to the most active features in a training loop and uploads them as learned information from local datasets. A pruning process is applied to the algorithm based on the validation set, which serves as a model accelerator. In the experiment, our model presents equal performances and higher saturating speed than the Federated Averaging method which reveals all the parameters of local models to the server when updating. We also demonstrate that the converging rates could be increased by introducing a pruning process.

cs.LG↗

arXiv2019

Floquet engineering to exotic topological phases in systems of cold atoms

Topological phases with a widely tunable number of edge modes have been extensively studied as a typical class of exotic states of matter with potentially important applications. Although several models have been shown to support such phases, they are not easy to realize in solid-state systems due to the complexity of various intervening factors. Inspired by the realization of synthetic spin-orbit coupling in a cold-atom system [Z. Wu {\it et al.}, Science \textbf{354}, 83 (2016)], we propose a periodic quenching scheme to realize large-topological-number phases with multiple edge modes in optical lattices. Via introducing the periodic quenching to the Raman lattice, it is found that a large number of edge modes can be induced in a controllable manner from the static topologically trivial system. Our result provides an experimentally accessible method to artificially synthesize and manipulate exotic topological phases with large topological numbers and multiple edge modes.

cond-mat.quant-gas↗

arXiv2019

Optimizing Pinning Control of Complex Dynamical Networks Based on Spectral Properties of Grounded Laplacian Matrices

Pinning control of a complex network aims at forcing the states of all nodes to track an external signal by controlling a small number of nodes in the network. In this paper, an algebraic graph-theoretic condition is introduced to optimize pinning control. When individual node dynamics and coupling strength of the network are given, the effectiveness of pinning scheme can be measured by the smallest eigenvalue of the grounded Laplacian matrix obtained by deleting the rows and columns corresponding to the pinned nodes from the Laplacian matrix of the network. The larger this smallest eigenvalue, the more effective the pinning scheme. Spectral properties of the smallest eigenvalue are analyzed using the network topology information, including the spectrum of the network Laplacian matrix, the minimal degree of uncontrolled nodes, the number of edges between the controlled node set and the uncontrolled node set, etc. The identified properties are shown effective for optimizing the pinning control strategy, as demonstrated by illustrative examples. Finally, for both scale-free and small-world networks, in order to maximize their corresponding smallest eigenvalues, it is better to pin the nodes with large degrees when the percentage of pinned nodes is relatively small, while it is better to pin nodes with small degrees when the percentage is relatively large. This surprising phenomenon can be explained by one of the theorems established.

math.OC↗

arXiv2019

Surpassing the rate-transmittance linear bound of quantum key distribution

Quantum key distribution (QKD offers a long-term solution to establish information-theoretically secure keys between two distant users. In practice, with a careful characterization of quantum sources and the decoy-state method, measure-device-independent quantum key distribution (MDI-QKD) provides secure key distribution. While short-distance fibre-based QKD has already been available for real-life implementation, the bottleneck of practical QKD lies on the limited transmission distance. Due to photon losses in transmission, it was believed that the key generation rate is bounded by a linear function of the channel transmittance, $O(η)$, without a quantum repeater, which puts an upper bound on the maximal secure transmission distance. Interestingly, a new phase-encoding MDI-QKD scheme, named twin-field QKD, has been suggested to beat the linear bound, while another variant, named phase-matching quantum key distribution (PM-QKD), has been proven to have a quadratic key-rate improvement, $O(\sqrtη)$. In reality, however, the intrinsic optical mode mismatch of independent lasers, accompanied by phase fluctuation and drift, impedes the successful experimental implementation of the new schemes. Here, we solve this problem with the assistance of the laser injection technique and the phase post-compensation method. In the experiment, the key rate surpasses the linear key-rate bound via 302 km and 402 km commercial-fibre channels, achieving a key rate over 4 orders of magnitude higher than the existing results in literature. Furthermore, with a 502 km ultralow-loss fibre, our system yields a secret key rate of 0.118 bps. We expect this new type of QKD schemes to become a new standard for future QKD.

quant-ph↗