arXiv subjects

Lei Zhou

Publications and source records attributed to Lei Zhou.

At least 109 records · Page 6Linked to original sources

arXiv2022

Trojan asteroids and co-orbital dust ring of Venus

The long-standing co-orbital asteroids have been thought to be the possible source of zodiacal dust ring around the orbit of Venus, but inconsistent conclusions on the orbital stability thus existence of Venus Trojans are found in literature. We present in this paper a systematic survey of the orbital stability of Venus Trojans, taking into account the dynamical influences from the general relativity and the Yarkovsky effect. The orbits of thousands of fictitious Venus Trojans are numerically simulated. Using the method of frequency analysis, their orbital stabilities and the dynamical mechanisms behind are described in detail. The influences of general relativity and of Yarkovsky effect, which were either neglected or oversimplified previously, are investigated by long-term numerical simulations. The stability maps on the $(a_0,i_0)$ plane and $(a_0,e_0)$ plane are depicted, and the most stable Venus Trojans are found to occupy the low-inclination horseshoe orbits with low eccentricities. The resonances that carve the fine structures in the stability map are determined. The general relativity decreases the stability of orbits by little but the Yarkovsky effect may drive nearly all Venus Trojans out of the Trojan region in a relatively short time. The Venus Trojans have poor orbital stability and cannot survive the age of the Solar system. The zodiacal dust ring found around the orbit of Venus is more likely a sporadic phenomenon, as the result of temporarily capture into the 1:1 mean motion resonance of dust particles produced probably from passing comets or asteroids, but not Venus Trojans.

astro-ph.EP↗

arXiv2022

Half a Dozen Real-World Applications of Evolutionary Multitasking, and More

Until recently, the potential to transfer evolved skills across distinct optimization problem instances (or tasks) was seldom explored in evolutionary computation. The concept of evolutionary multitasking (EMT) fills this gap. It unlocks a population's implicit parallelism to jointly solve a set of tasks, hence creating avenues for skills transfer between them. Despite it being early days, the idea of EMT has begun to show promise in a range of real-world applications. In the backdrop of recent advances, the contribution of this paper is twofold. First, a review of several application-oriented explorations of EMT in the literature is presented; the works are assimilated into half a dozen broad categories according to their respective application domains. Each of these six categories elaborates fundamental motivations to multitask, and contains a representative experimental study (referred from the literature). Second, a set of recipes is provided showing how problem formulations of general interest, those that cut across different disciplines, could be transformed in the new light of EMT. Our discussions emphasize the many practical use-cases of EMT, and is intended to spark future research towards crafting novel algorithms for real-world deployment.

cs.NE↗

arXiv2022

EnclaveTree: Privacy-preserving Data Stream Training and Inference Using TEE

The classification service over a stream of data is becoming an important offering for cloud providers, but users may encounter obstacles in providing sensitive data due to privacy concerns. While Trusted Execution Environments (TEEs) are promising solutions for protecting private data, they remain vulnerable to side-channel attacks induced by data-dependent access patterns. We propose a Privacy-preserving Data Stream Training and Inference scheme, called EnclaveTree, that provides confidentiality for user's data and the target models against a compromised cloud service provider. We design a matrix-based training and inference procedure to train the Hoeffding Tree (HT) model and perform inference with the trained model inside the trusted area of TEEs, which provably prevent the exploitation of access-pattern-based attacks. The performance evaluation shows that EnclaveTree is practical for processing the data streams with small or medium number of features. When there are less than 63 binary features, EnclaveTree is up to ${\thicksim}10{\times}$ and ${\thicksim}9{\times}$ faster than naïve oblivious solution on training and inference, respectively.

cs.CR↗

arXiv2022

Control Co-design of Actively Controlled Lightweight Structures for High-acceleration Precision Motion Systems

Precision motion stages are an essential part of a wide range of manufacturing equipment, and their motion performance are critical to the quality and throughput of the systems. The drastically increasing demand for higher manufacturing throughput in various processes necessities the development of next-generation motion systems with reduced moving weight and high control bandwidth. However, the reduction of moving stage's weight can lower the stage's structural resonance frequencies, making the hardware dynamics and controller design problem strongly coupled. Aiming at this challenge, this paper proposes a new formulation of nested hardware and control co-design framework for precision motion stages. The proposed framework explicitly optimizes the closed-loop control bandwidth with guaranteed robustness, and explicitly considers the limits in the physical system. Two case studies, including a motivating example using lumped-parameter mechanical system and a finite-element-simulated lightweight motion stage, are being used to evaluate the effectiveness of the proposed nested CCD framework. Simulation results show that the proposed nested CCD framework has 42\% of weight reduction and 28\% bandwidth improvement compared with a sequential design baseline, which demonstrates the efficacy of the proposed approach.

eess.SY↗

arXiv2022

Self-Sensing Hysteresis-Type Bearingless Motor

Bearingless motors use a single stator assembly to apply torque and magnetic suspension forces on the rotor, making these machines compact with frictionless operation and thus well suited to high-speed applications. One major challenge that prevents wide usage of bearingless motors is the need for air-gap position sensors, which are typically expensive. Here we present a method to estimate the radial position of a hysteresis-type bearingless motor using the inductance variation of the stator coils amplified by an injected high-frequency signal. We have carried out finite element (FE) simulations to demonstrate its feasibility, and have constructed a prototype self-sensing bearingless motor for experimental validations.

eess.SY↗

arXiv2022

Lung Swapping Autoencoder: Learning a Disentangled Structure-texture Representation of Chest Radiographs

Well-labeled datasets of chest radiographs (CXRs) are difficult to acquire due to the high cost of annotation. Thus, it is desirable to learn a robust and transferable representation in an unsupervised manner to benefit tasks that lack labeled data. Unlike natural images, medical images have their own domain prior; e.g., we observe that many pulmonary diseases, such as the COVID-19, manifest as changes in the lung tissue texture rather than the anatomical structure. Therefore, we hypothesize that studying only the texture without the influence of structure variations would be advantageous for downstream prognostic and predictive modeling tasks. In this paper, we propose a generative framework, the Lung Swapping Autoencoder (LSAE), that learns factorized representations of a CXR to disentangle the texture factor from the structure factor. Specifically, by adversarial training, the LSAE is optimized to generate a hybrid image that preserves the lung shape in one image but inherits the lung texture of another. To demonstrate the effectiveness of the disentangled texture representation, we evaluate the texture encoder $Enc^t$ in LSAE on ChestX-ray14 (N=112,120), and our own multi-institutional COVID-19 outcome prediction dataset, COVOC (N=340 (Subset-1) + 53 (Subset-2)). On both datasets, we reach or surpass the state-of-the-art by finetuning $Enc^t$ in LSAE that is 77% smaller than a baseline Inception v3. Additionally, in semi-and-self supervised settings with a similar model budget, $Enc^t$ in LSAE is also competitive with the state-of-the-art MoCo. By "re-mixing" the texture and shape factors, we generate meaningful hybrid images that can augment the training set. This data augmentation method can further improve COVOC prediction performance. The improvement is consistent even when we directly evaluate the Subset-1 trained model on Subset-2 without any fine-tuning.

eess.IV↗

arXiv2021

On Some Modal Implications of the Dynamic Mode Decomposition Through the Lens of a Subcritical Prism Wake

The Dynamic Mode Decomposition (DMD) is a Koopman-based algorithm that straightforwardly isolates individual mechanisms from the compound morphology of direct measurement. However, many may be perplexed by the messages the DMD structures carry. This work investigates the modal implications of the DMD/Koopman modes through the prototypical subcritical free-shear flow over a square prism. It selected and analysed the fluid mechanics and phenomenology of the ten most dominant modes. The results showed that the reduced-order description is morphologically accurate and physically insightful. Mode 1 renders the mean-field. Modes 2 depicts the roll-up of the Strouhal vortex. Mode 3 delineates the Bloor-Gerrard vortex resulting from the Kelvin-Helmholtz instability inside shear layers, its superposition onto the Strouhal vortex, and the concurrent flow entrainment. Modes 4, 5, 7, 8, and 9 portray the harmonic excitation. Modes 6 and 10 describe the low-frequency shedding of turbulent separation bubbles (TSBs) and turbulence production, respectively, which contribute to the beating phenomenon in the lift time history and the flapping motion of shear layers. Finally, this work demonstrates the capability of the DMD in providing insights into similar fluid problems. It also serves as an excellent reference for an array of other nonlinear systems.

physics.flu-dyn↗

arXiv2021

Perfect anomalous reflectors at optical frequencies

Reflecting light to a pre-determined non-specular direction is an important ability of metasurfaces, which is the basis for a wide range of applications (e.g., beam steering/splitting and imaging). However, anomalous reflection with 100% efficiency has not been achieved at optical frequencies in conventional metasurfaces, due to losses and/or insufficient nonlocal control of light waves. Here, we propose a new type of all-dielectric quasi-three-dimensional subwavelength structures, consisting of multilayer films and specifically designed meta-gratings, to achieve perfect anomalous reflections at optical frequencies. A complex multiple scattering process was stimulated by effectively coupling different Bloch waves and propagating waves in the proposed meta-system, thus offering the whole meta-system the desired nonlocal control on light waves required to achieve perfect anomalous reflections. Two perfect anomalous reflectors were designed to reflect normally incident 1550 nm light to the 40° and 75° directions with absolute efficiencies higher than 99%, and were subsequently fabricated and experimentally demonstrated to exhibit efficiencies 98% and 88%, respectively. Our results pave the way towards realizing optical meta-devices with desired high efficiencies in realistic applications.

physics.optics↗

arXiv2021

Implementation of a 46-node quantum metropolitan area network

Quantum key distribution (QKD) enables secure key exchanges between two remote users. The ultimate goal of secure communication is to establish a global quantum network. The existing field tests suggest that quantum networks are feasible. To achieve a practical quantum network, we need to overcome several challenges, including realising versatile topologies for large scales, simple network maintenance, extendable configuration, and robustness to node failures. To this end, we present a field operation of a quantum metropolitan-area network with 46 nodes and show that all these challenges can be overcome with cutting-edge quantum technologies. In particular, we realise different topological structures and continuously run the network for 31 months, by employing standard equipment for network maintenance with an extendable configuration. We realise QKD pairing and key management with a sophisticated key control center. In this implementation, the final keys have been used for secure communication such as real-time voice telephone, text messaging, and file transmission with one-time pad encryption, which can support 11 pairs of users to make audio calls simultaneously. Combined with inter-city quantum backbone and ground-satellite links, our metropolitan implementation paves the way toward a global quantum network.

quant-ph↗

arXiv2021

Tadpole type motion of charged dust in the Lagrange problem with planet Jupiter

We investigate the dynamics of charged dust interacting with the interplanetary magnetic field in a Parker spiral type model and subject to the solar wind and Poynting-Robertson effect in the vicinity of the 1:1 mean motion resonance with planet Jupiter. We estimate the shifts of the location of the minimum libration amplitude solutions close to the location of the L4 and L5 points of the classical - gravitational - problem and provide the extension of the 'librational regimes of motion' and the width of the resonance in dependency of the nongravitational parameters related to the dust grain size and surface potential of the particles. Our study is based on numerical simulations in the framework of the spatial, elliptic restricted three-body problem and semi-analytical estimates obtained by averaging of Gauss' planetary equations of motion.

astro-ph.EP↗

arXiv2021

Self-Guided Curriculum Learning for Neural Machine Translation

In the field of machine learning, the well-trained model is assumed to be able to recover the training labels, i.e. the synthetic labels predicted by the model should be as close to the ground-truth labels as possible. Inspired by this, we propose a self-guided curriculum strategy to encourage the learning of neural machine translation (NMT) models to follow the above recovery criterion, where we cast the recovery degree of each training example as its learning difficulty. Specifically, we adopt the sentence level BLEU score as the proxy of recovery degree. Different from existing curricula relying on linguistic prior knowledge or third-party language models, our chosen learning difficulty is more suitable to measure the degree of knowledge mastery of the NMT models. Experiments on translation benchmarks, including WMT14 English$\Rightarrow$German and WMT17 Chinese$\Rightarrow$English, demonstrate that our approach can consistently improve translation performance against strong baseline Transformer.

cs.CL↗

arXiv2021

Learning to Match Features with Seeded Graph Matching Network

Matching local features across images is a fundamental problem in computer vision. Targeting towards high accuracy and efficiency, we propose Seeded Graph Matching Network, a graph neural network with sparse structure to reduce redundant connectivity and learn compact representation. The network consists of 1) Seeding Module, which initializes the matching by generating a small set of reliable matches as seeds. 2) Seeded Graph Neural Network, which utilizes seed matches to pass messages within/across images and predicts assignment costs. Three novel operations are proposed as basic elements for message passing: 1) Attentional Pooling, which aggregates keypoint features within the image to seed matches. 2) Seed Filtering, which enhances seed features and exchanges messages across images. 3) Attentional Unpooling, which propagates seed features back to original keypoints. Experiments show that our method reduces computational and memory complexity significantly compared with typical attention-based networks while competitive or higher performance is achieved.

cs.CV↗

arXiv2021

Active spintronic-metasurface terahertz emitters with tunable chirality

The ability to manipulate the electric-field vector of broadband terahertz waves is essential for applications of terahertz technologies in many areas, and can open up new possibilities for nonlinear terahertz spectroscopy and coherent control. Here, we propose a novel laser-driven terahertz emitter, consisting of metasurface-patterned magnetic multilayer heterostructures. Such hybrid terahertz emitters can combine the advantages of spintronic emitters for being ultrabroadband, efficient and flexible, as well as those of metasurfaces for the unique capability to manipulate terahertz waves with high precision and degree of freedom. Taking a stripe-patterned metasurface as an example, we demonstrate the generation of broadband terahertz waves with tunable chirality. Based on experimental and theoretical studies, the interplay between the laser-induced spintronic-origin currents and the metasurface-induced transient charges/currents are investigated, revealing the strong influence on the device functionality originated from both the light-matter interactions in individual metasurface units and the dynamic coupling between them. Our work not only offers a flexible, reliable and cost-effective solution for chiral terahertz wave generation and manipulation, but also opens a new pathway to metasurface-tailored spintronic devices for efficient vector-control of electromagnetic waves in the terahertz regime.

physics.optics↗

arXiv2021

On the formation of terrestrial planets between two massive planets: The case of 55 Cancri

Considering the huge computational resources required by smoothed particle hydrodynamics (SPH) simulations and the overestimation of post-collision materials from perfect merging, we develop a statistical method to deal with collisions during the formation of planetary systems by introducing random material loss. In this method the mass and water content lost by the sole outcome from every merger vary randomly within a range dependent on the total mass and water content of colliding bodies. The application of the random loss method to the planet formation in the solar system shows a good consistency with existing SPH results. We also apply this method to the extrasolar planetary system 55 Cancri which hosts (at least) five planets and study the formation of terrestrial planets between the outermost two planets. A disk with 500 Mars mass embryos in dynamically cold orbits before the late-stage accretion phase is assumed. Scenarios with different amounts of planetary embryos and different loss parameters are adopted in our simulations. The statistical result from hundreds of simulations shows that an Earth-like planet with water inventory of roughly 6 Earth ocean could form between 55 Cnc f and d. It may reside between 1.0 and 2.6 AU but the most likely region extends from 1.5 to 2.1 AU. Thus the probability of this planet being in the potentially habitable zone (0.59--1.43 AU) is relatively low, only around 10\%. Planets 55 Cnc f and d could also be shaped and gain some water from giant impacts and consequently the orbits of them may also change accordingly.

astro-ph.EP↗

arXiv2021

The Stable Marriage Problem: an Interdisciplinary Review from the Physicist's Perspective

We present a fascinating model that has lately caught attention among physicists working in complexity related fields. Though it originated from mathematics and later from economics, the model is very enlightening in many aspects that we shall highlight in this review. It is called The Stable Marriage Problem (though the marriage metaphor can be generalized to many other contexts), and it consists of matching men and women, considering preference-lists where individuals express their preference over the members of the opposite gender. This problem appeared for the first time in 1962 in the seminal paper of Gale and Shapley and has aroused interest in many fields of science, including economics, game theory, computer science, etc. Recently it has also attracted many physicists who, using the powerful tools of statistical mechanics, have also approached it as an optimization problem. Here we present a complete overview of the Stable Marriage Problem emphasizing its multidisciplinary aspect, and reviewing the key results in the disciplines that it has influenced most. We focus, in particular, in the old and recent results achieved by physicists, finally introducing two new promising models inspired by the philosophy of the Stable Marriage Problem. Moreover, we present an innovative reinterpretation of the problem, useful to highlight the revolutionary role of information in the contemporary economy.

physics.soc-ph↗

arXiv2021

Cascaded Feature Warping Network for Unsupervised Medical Image Registration

Deformable image registration is widely utilized in medical image analysis, but most proposed methods fail in the situation of complex deformations. In this paper, we pre-sent a cascaded feature warping network to perform the coarse-to-fine registration. To achieve this, a shared-weights encoder network is adopted to generate the feature pyramids for the unaligned images. The feature warping registration module is then used to estimate the deformation field at each level. The coarse-to-fine manner is implemented by cascading the module from the bottom level to the top level. Furthermore, the multi-scale loss is also introduced to boost the registration performance. We employ two public benchmark datasets and conduct various experiments to evaluate our method. The results show that our method outperforms the state-of-the-art methods, which also demonstrates that the cascaded feature warping network can perform the coarse-to-fine registration effectively and efficiently.

cs.CV↗

arXiv2021

PointDSC: Robust Point Cloud Registration using Deep Spatial Consistency

Removing outlier correspondences is one of the critical steps for successful feature-based point cloud registration. Despite the increasing popularity of introducing deep learning methods in this field, spatial consistency, which is essentially established by a Euclidean transformation between point clouds, has received almost no individual attention in existing learning frameworks. In this paper, we present PointDSC, a novel deep neural network that explicitly incorporates spatial consistency for pruning outlier correspondences. First, we propose a nonlocal feature aggregation module, weighted by both feature and spatial coherence, for feature embedding of the input correspondences. Second, we formulate a differentiable spectral matching module, supervised by pairwise spatial compatibility, to estimate the inlier confidence of each correspondence from the embedded features. With modest computation cost, our method outperforms the state-of-the-art hand-crafted and learning-based outlier rejection approaches on several real-world datasets by a significant margin. We also show its wide applicability by combining PointDSC with different 3D local descriptors.

cs.CV↗

arXiv2021

Goal-Oriented Gaze Estimation for Zero-Shot Learning

Zero-shot learning (ZSL) aims to recognize novel classes by transferring semantic knowledge from seen classes to unseen classes. Since semantic knowledge is built on attributes shared between different classes, which are highly local, strong prior for localization of object attribute is beneficial for visual-semantic embedding. Interestingly, when recognizing unseen images, human would also automatically gaze at regions with certain semantic clue. Therefore, we introduce a novel goal-oriented gaze estimation module (GEM) to improve the discriminative attribute localization based on the class-level attributes for ZSL. We aim to predict the actual human gaze location to get the visual attention regions for recognizing a novel object guided by attribute description. Specifically, the task-dependent attention is learned with the goal-oriented GEM, and the global image features are simultaneously optimized with the regression of local attribute features. Experiments on three ZSL benchmarks, i.e., CUB, SUN and AWA2, show the superiority or competitiveness of our proposed method against the state-of-the-art ZSL methods. The ablation analysis on real gaze data CUB-VWSW also validates the benefits and accuracy of our gaze estimation module. This work implies the promising benefits of collecting human gaze dataset and automatic gaze estimation algorithms on high-level computer vision tasks. The code is available at https://github.com/osierboy/GEM-ZSL.

cs.CV↗