arXiv subjects

Lin Wu

Publications and source records attributed to Lin Wu.

At least 73 records · Page 4Linked to original sources

arXiv2022

Ultra-Low-Frequency Radio Astronomy Observations from a Selenocentric Orbit: first results of the Longjiang-2 experiment

This paper introduces the first results of observations with the Ultra-Long-Wavelength (ULW) -- Low Frequency Interferometer and Spectrometer (LFIS) on board the selenocentric satellite Longjiang-2. We present a brief description of the satellite and focus on the LFIS payload. The in-orbit commissioning confirmed a reliable operational status of the instrumentation. We also present results of a transition observation, which offers unique measurements on several novel aspects. We estimate the RFI suppression required for such a radio astronomy instrumentation at the Moon distances from Earth to be of the order of 80 dB. We analyse a method of separating Earth- and satellite-originated radio frequency interference (RFI). It is found that the RFI level at frequencies lower than a few MHz is smaller than the receiver noise floor.

astro-ph.IM↗

arXiv2022

Jacobian Norm with Selective Input Gradient Regularization for Improved and Interpretable Adversarial Defense

Deep neural networks (DNNs) are known to be vulnerable to adversarial examples that are crafted with imperceptible perturbations, i.e., a small change in an input image can induce a mis-classification, and thus threatens the reliability of deep learning based deployment systems. Adversarial training (AT) is often adopted to improve robustness through training a mixture of corrupted and clean data. However, most of AT based methods are ineffective in dealing with transferred adversarial examples which are generated to fool a wide spectrum of defense models, and thus cannot satisfy the generalization requirement raised in real-world scenarios. Moreover, adversarially training a defense model in general cannot produce interpretable predictions towards the inputs with perturbations, whilst a highly interpretable robust model is required by different domain experts to understand the behaviour of a DNN. In this work, we propose a novel approach based on Jacobian norm and Selective Input Gradient Regularization (J-SIGR), which suggests the linearized robustness through Jacobian normalization and also regularizes the perturbation-based saliency maps to imitate the model's interpretable predictions. As such, we achieve both the improved defense and high interpretability of DNNs. Finally, we evaluate our method across different architectures against powerful adversarial attacks. Experiments demonstrate that the proposed J-SIGR confers improved robustness against transferred adversarial attacks, and we also show that the predictions from the neural network are easy to interpret.

cs.LG↗

arXiv2022

Exploring variational quantum eigensolver ansatzes for the long-range XY model

Finding the ground state energy and wavefunction of a quantum many-body system is a key problem in quantum physics and chemistry. We study this problem for the long-range XY model by using the variational quantum eigensolver (VQE) algorithm. We consider VQE ansatzes with full and linear entanglement structures consisting of different building gates: the CNOT gate, the controlled-rotation (CRX) gate, and the two-qubit rotation (TQR) gate. We find that the full-entanglement CRX and TQR ansatzes can sufficiently describe the ground state energy of the long-range XY model. In contrast, only the full-entanglement TQR ansatz can represent the ground state wavefunction with a fidelity close to one. In addition, we find that instead of using full-entanglement ansatzes, restricted-entanglement ansatzes where entangling gates are applied only between qubits that are a fixed distance from each other already suffice to give acceptable solutions. Using the entanglement entropy to characterize the expressive powers of the VQE ansatzes, we show that the full-entanglement TQR ansatz has the highest expressive power among them.

quant-ph↗

arXiv2022

Pseudo-Pair based Self-Similarity Learning for Unsupervised Person Re-identification

Person re-identification (re-ID) is of great importance to video surveillance systems by estimating the similarity between a pair of cross-camera person shorts. Current methods for estimating such similarity require a large number of labeled samples for supervised training. In this paper, we present a pseudo-pair based self-similarity learning approach for unsupervised person re-ID without human annotations. Unlike conventional unsupervised re-ID methods that use pseudo labels based on global clustering, we construct patch surrogate classes as initial supervision, and propose to assign pseudo labels to images through the pairwise gradient-guided similarity separation. This can cluster images in pseudo pairs, and the pseudos can be updated during training. Based on pseudo pairs, we propose to improve the generalization of similarity function via a novel self-similarity learning:it learns local discriminative features from individual images via intra-similarity, and discovers the patch correspondence across images via inter-similarity. The intra-similarity learning is based on channel attention to detect diverse local features from an image. The inter-similarity learning employs a deformable convolution with a non-local block to align patches for cross-image similarity. Experimental results on several re-ID benchmark datasets demonstrate the superiority of the proposed method over the state-of-the-arts.

cs.CV↗

arXiv2022

Learning Resolution-Adaptive Representations for Cross-Resolution Person Re-Identification

The cross-resolution person re-identification (CRReID) problem aims to match low-resolution (LR) query identity images against high resolution (HR) gallery images. It is a challenging and practical problem since the query images often suffer from resolution degradation due to the different capturing conditions from real-world cameras. To address this problem, state-of-the-art (SOTA) solutions either learn the resolution-invariant representation or adopt super-resolution (SR) module to recover the missing information from the LR query. This paper explores an alternative SR-free paradigm to directly compare HR and LR images via a dynamic metric, which is adaptive to the resolution of a query image. We realize this idea by learning resolution-adaptive representations for cross-resolution comparison. Specifically, we propose two resolution-adaptive mechanisms. The first one disentangles the resolution-specific information into different sub-vectors in the penultimate layer of the deep neural networks, and thus creates a varying-length representation. To better extract resolution-dependent information, we further propose to learn resolution-adaptive masks for intermediate residual feature blocks. A novel progressive learning strategy is proposed to train those masks properly. These two mechanisms are combined to boost the performance of CRReID. Experimental results show that the proposed method is superior to existing approaches and achieves SOTA performance on multiple CRReID benchmarks.

cs.CV↗

arXiv2022

Fiber spectrum analyzer based on planar waveguide array aligned to a camera without lens

We propose and experimentally demonstrate a fiber spectrum analyzer based on a planar waveguide chip butt-coupled with an input fiber and aligned to a standard camera without any free-space optical elements. The chip consists of a single-mode waveguide to connect with the fiber, a beam broadening area, and a waveguide array in which the lengths of the waveguides are designed for both wavelength separation and beam focusing. The facet of the chip is diced open so that the outputs of the array form a near-field emitter. The far field are calculated by the Rayleigh-Sommerfeld diffraction integral. We show that the chip can provide a focal depth on the millimeter scale, allowing relaxed alignment to the camera without any fine-positioning stage. Two devices with 120 and 220 waveguides are fabricated on the polymer waveguide platform. The measured spectral width are 0.63 nm and 0.42 nm, respectively. This simple and practical approach may lead to the development of a spectrum analyzer for fiber that is easily mountable to any commercial camera, thereby avoiding the complication for customized detectors as well as electronic circuits afterwards.

physics.optics↗

arXiv2022

Designing light-element materials with large effective spin-orbit coupling

Spin-orbit coupling (SOC), the core of numerous condensed-matter phenomena such as nontrivial band gap, magnetocrystalline anisotropy, etc, is generally considered to be appreciable only in heavy elements, detrimental to the synthetization and application of functional materials. Therefore, amplifying the SOC effect in light elements is of great importance. Here, focusing on 3d and 4d systems, we demonstrate that the interplay between crystal symmetry and electron correlation can dramatically enhance the SOC effect in certain partially occupied orbital multiplets, through the self-consistently reinforced orbital polarization as a pivot. We then provide design principles and comprehensive databases, in which we list all the Wyckoff positions and site symmetries, in all two-dimensional (2D) and three-dimensional crystals that potentially have such enhanced SOC effect. As an important demonstration, we predict nine material candidates from our selected 2D material pool as high-temperature quantum anomalous Hall insulators with large nontrivial band gaps of hundreds of meV. Our work provides an efficient and straightforward way to predict promising SOC-active materials, releasing the burden of requiring heavy elements for next-generation spin-orbitronic materials and devices.

cond-mat.mtrl-sci↗

arXiv2022

Multi-modal Visual Place Recognition in Dynamics-Invariant Perception Space

Visual place recognition is one of the essential and challenging problems in the fields of robotics. In this letter, we for the first time explore the use of multi-modal fusion of semantic and visual modalities in dynamics-invariant space to improve place recognition in dynamic environments. We achieve this by first designing a novel deep learning architecture to generate the static semantic segmentation and recover the static image directly from the corresponding dynamic image. We then innovatively leverage the spatial-pyramid-matching model to encode the static semantic segmentation into feature vectors. In parallel, the static image is encoded using the popular Bag-of-words model. On the basis of the above multi-modal features, we finally measure the similarity between the query image and target landmark by the joint similarity of their semantic and visual codes. Extensive experiments demonstrate the effectiveness and robustness of the proposed approach for place recognition in dynamic environments.

cs.CV↗

arXiv2021

Controlling plexcitonic strong coupling via multidimensional hotspot nanoengineering

Plexcitonic strong coupling has ushered in an era of room-temperature quantum electrodynamics that is achievable at the nanoscale, with potential applications ranging from high-precision single-molecule spectroscopy to quantum technologies functional under ambient conditions. Realizing these applications on an industrial scale requires scalable and mass-producible plasmonic cavities that provide ease of access and control for quantum emitters. Via a rational selection of substrates and the canonical gold bowtie nanoantenna, we propose a novel design strategy for multidimensional engineering of nanocavity antenna-mode hotspots, which facilitates their elevation to the top of the nanobowtie gap and provides a field enhancement of ~500 fold (a 1.6-fold increase compared to a conventional nanobowtie-on-glass cavity at the bottom of the nanobowtie gap). We discuss the formation mechanism for such antenna modes using different material substrates from the perspective of charge carrier motion, and analyze their sensitivity to the geometrical parameters of the device. The advantages of these antenna modes, particularly in view of their dominantly in-plane polarized near-fields, are further elaborated in a spatiotemporal study of plexcitonic strong coupling involving single emitters and layered ensembles thereof, which reveals ultrafast quantum dynamics dependent on both the substrate and nanobowtie geometry, as well as the potential for applications related to 2D materials whose excitonic dipoles are typically oriented in-plane. The conceptual discovery of this substrate-enabled antenna-mode nanoengineering could readily be extended to tailor hotspots in other plasmonic platforms, and we anticipate that this work could inspire a wide range of novel research directions from photoluminescence spectroscopy and sensing to the design of quantum logic gates and systems for long-range energy transfer.

physics.optics↗

arXiv2021

Imaging sensitivity of a linear interferometer array on lunar orbit

Ground-based observation at frequencies below 30 MHz is hindered by the ionosphere of the Earth and radio frequency interference. To map the sky at these low frequencies, we have proposed the Discovering the Sky at the Longest wavelength mission (DSL, also known as the "Hongmeng" mission, which means "Primordial Universe" in Chinese) concept, which employs a linear array of micro-satellites orbiting the Moon. Such an array can be deployed to the lunar orbit by a single rocket launch, and it can make interferometric observations achieving good angular resolutions despite the small size of the antennas. However, it differs from the conventional ground-based interferometer array or even the previous orbital interferometers in many aspects, new data-processing methods need to be developed. In this work, we make a series of simulations to assess the imaging quality and sensitivity of such an array. We start with an input sky model and a simple orbit model, generate mock interferometric visibilities, and then reconstruct the sky map. We consider various observational effects and practical issues, such as the system noise, antenna response, and Moon blockage. Based on the quality of the recovered image, we quantify the imaging capability of the array for different satellite numbers and array configurations. For the first time, we make practical estimates of the point source sensitivity for such a lunar orbit array, and predict the expected number of detectable sources for the mission. Depending on the radio source number distribution which is still very uncertain at these frequencies, the proposed mission can detect $10^2 \sim 10^4$ sources during its operation.

astro-ph.IM↗

arXiv2021

Cross-speaker Emotion Transfer Based on Speaker Condition Layer Normalization and Semi-Supervised Training in Text-To-Speech

In expressive speech synthesis, there are high requirements for emotion interpretation. However, it is time-consuming to acquire emotional audio corpus for arbitrary speakers due to their deduction ability. In response to this problem, this paper proposes a cross-speaker emotion transfer method that can realize the transfer of emotions from source speaker to target speaker. A set of emotion tokens is firstly defined to represent various categories of emotions. They are trained to be highly correlated with corresponding emotions for controllable synthesis by cross-entropy loss and semi-supervised training strategy. Meanwhile, to eliminate the down-gradation to the timbre similarity from cross-speaker emotion transfer, speaker condition layer normalization is implemented to model speaker characteristics. Experimental results show that the proposed method outperforms the multi-reference based baseline in terms of timbre similarity, stability and emotion perceive evaluations.

eess.AS↗

arXiv2021

Suppressing Decoherence in Quantum Plasmonic Systems by Spectral Hole Burning Effect

Quantum plasmonic systems suffer from significant decoherence due to the intrinsically large dissipative and radiative dampings. Based on our quantum simulations via a quantum tensor network algorithm, we numerically demonstrate the mitigation of this restrictive drawback by hybridizing a plasmonic nanocavity with an emitter ensemble with inhomogeneously-broadened transition frequencies. By burning two narrow spectral holes in the spectral density of the emitter ensemble, the coherent time of Rabi oscillation for the hybrid system is increased tenfold. With the suppressed decoherence, we move one step further in bringing plasmonic systems into practical quantum applications.

quant-ph↗

arXiv2021

Symmetry-enforced Band Nodes in 230 Space Groups

Crystallographic symmetries enforcing band touchings (BTs) in the Brillouin zone (BZ) have been utilized to classify and predict the topological semimetals. Though the early proposed topological semimetals contain isolated nodal points in the BZ, the proposed nodal line semimetals later could host various structures of several nodal lines/loops: nodal chains, nodal nets or Hopf-links, etc. In this work, using compatibility relations, we first list all possible high symmetry lines (HSLs) that can be nodal lines itself, high symmetry planes (HSPLs) that can host nodal loops, high symmetry planes (HSPLs) that are nodal surfaces for all 230 SGs, with spin-orbit coupling and time-reversal symmetry considered or not. We then show how to diagnose a nodal loop from the band crossing in an HSL, or nodal line/surface from irreducible representation (irrep) of an high-symmetry point (HSP), while the rest cases correspond to nodal points. Among our results, those essential cases, for which the nodal points/lines/loops/surfaces must exist, are highlighted since they are promising for the realizations of (nearly) ideal nodal point/line/loop/surface semimetals, as well as systems with flexible tunability owning fixed structure of topological nodal points/lines/loops/surfaces. Based on our results, SGs allowing Hopf-link structure with one straight nodal line threading a nodal loop, or two nesting nodal loops lying in two respective high symmetry planes, are highlighted, with the predicted materials being B$_5$Pb$_2$IO$_9$ in SG 34 and SrAl$_2$Au$_3$ in SG 62, respectively. Our exhaustive results could serve as a useful guide for efficiently predicting and designing materials or artificial systems owning exotic geometric nodal structures of energy bands simply based on structure symmetries.

cond-mat.mtrl-sci↗

arXiv2020

2D-Material-Assisted Bistable Switching of Gap Plasmons Disclosed By Femtosecond Pulse Scattering Spectra

Nanosphere-on-mirror plasmonic antennas, each having a monolayer graphene or MoS2 sheet in the gap, were pumped with a femtosecond laser. Abrupt turnings in the scattering linewidth and peak intensity trends as the laser power changed were experimentally observed. Theoretical modelling of dynamic plasmon evolvement attributes the turning to transitioning between two plasmon states, with a universal switching threshold of four-wave mixing efficiency around 0.14%. This bistability is rendered by a strong feedback from the nonlinear gap current to the gap plasmons, and involvement of both four and high-order wave mixing. This work reveals a pathway to making energy efficient nonlinear plasmonic elements.

physics.optics↗

arXiv2020

Discovering the Sky at the Longest wavelengths with a lunar orbit array

Due to ionosphere absorption and the interference by natural and artificial radio emissions, astronomical observation from the ground becomes very difficult at the wavelengths of decametre or longer, which we shall refer as the ultralong wavelengths. This unexplored part of electromagnetic spectrum has the potential of great discoveries, notably in the study of cosmic dark ages and dawn, but also in heliophysics and space weather, planets and exoplanets, cosmic ray and neutrinos, pulsar and interstellar medium (ISM), extragalactic radio sources, and so on. The difficulty of the ionosphere can be overcome by space observation, and the Moon can shield the radio frequency interferences (RFIs) from the Earth. A lunar orbit array can be a practical first step of opening up the ultralong wave band. Compared with a lunar surface observatory on the far side, the lunar orbit array is simpler and more economical, as it does not need to make the risky and expensive landing, can be easily powered with solar energy, and the data can be transmitted back to the Earth when it is on the near-side part of the orbit. Here I describe the Discovering Sky at the Longest wavelength (DSL) project, which will consist of a mother satellite and 6~9 daughter satellites, flying on the same circular orbit around the Moon, and forming a linear interferometer array. The data are collected by the mother satellite which computes the interferometric cross-correlations (visibilities) and transmits the data back to the Earth. The whole array can be deployed on the lunar orbit with a single rocket launch. The project is under intensive study in China.

astro-ph.IM↗

arXiv2020

Reconfigurable photon sources based on quantum plexcitonic systems

A single photon in a strongly nonlinear cavity is able to block the transmission of the second photon, thereby converting incident coherent light into anti-bunched light, which is known as photon blockade effect. On the other hand, photon anti-pairing, where only the entry of two photons is blocked and the emission of bunches of three or more photons is allowed, is based on an unconventional photon blockade mechanism due to destructive interference of two distinct excitation pathways. We propose quantum plexcitonic systems with moderate nonlinearity to generate both anti-bunched and anti-paired photons. The proposed plexitonic systems benefit from subwavelength field localizations that make quantum emitters spatially distinguishable, thus enabling a reconfigurable photon source between anti-bunched and anti-paired states via tailoring the energy bands. For a realistic nanoprism plexitonic system, two schemes of reconfiguration are suggested: (i) the chemical means by partially changing the type of the emitters; or (ii) the optical approach by rotating the polarization angle of the incident light to tune the coupling rate of the emitters. These results pave the way to realize reconfigurable nonclassical photon sources in a simple quantum plexcitonic platform with readily accessible experimental conditions.

quant-ph↗

arXiv2020

Unsupervised Domain Adaptive Object Detection using Forward-Backward Cyclic Adaptation

We present a novel approach to perform the unsupervised domain adaptation for object detection through forward-backward cyclic (FBC) training. Recent adversarial training based domain adaptation methods have shown their effectiveness on minimizing domain discrepancy via marginal feature distributions alignment. However, aligning the marginal feature distributions does not guarantee the alignment of class conditional distributions. This limitation is more evident when adapting object detectors as the domain discrepancy is larger compared to the image classification task, e.g. various number of objects exist in one image and the majority of content in an image is the background. This motivates us to learn domain invariance for category level semantics via gradient alignment. Intuitively, if the gradients of two domains point in similar directions, then the learning of one domain can improve that of another domain. To achieve gradient alignment, we propose Forward-Backward Cyclic Adaptation, which iteratively computes adaptation from source to target via backward hopping and from target to source via forward passing. In addition, we align low-level features for adapting holistic color/texture via adversarial training. However, the detector performs well on both domains is not ideal for target domain. As such, in each cycle, domain diversity is enforced by maximum entropy regularization on the source domain to penalize confident source-specific learning and minimum entropy regularization on target domain to intrigue target-specific learning. Theoretical analysis of the training process is provided, and extensive experiments on challenging cross-domain object detection datasets have shown the superiority of our approach over the state-of-the-art.

cs.CV↗

arXiv2020

Exhaustive List of Topological Hourglass Band Crossings in 230 Space Groups

Topological semimetals with band crossings (BCs) near the Fermi level have attracted intense research activities in the past several years. Among various BCs, those enforced by an hourglass-like connectivity pattern, which are just located at the vertex in the neck of an hourglass and thus called hourglass BCs (HBCs), show interesting topological properties and are intimately related with the space group symmetry. Through checking compatibility relations in the Brillouin zone (BZ), we list all possible HBCs for all 230 space groups by identifying positions of HBCs as well as the compatibility relations related with the HBCs.The HBCs can be coexisting with conventional topological BCs such as Dirac andWeyl fermions and based on our exhaustive list, the dimensionality and degeneracy of the HBCs can be quickly identified. It is also found that the HBCs can be classified into two categories: one contains essential HBCs which are guaranteed to exist, while the HBCs in the other category may be tuned to disappear. Our results can help in efficiently predicting hourglass semimetals combined with first-principles calculations as well as studying transitions among various topological crystalline phases.

cond-mat.mtrl-sci↗