arXiv subjects

Xin Wang

Publications and source records attributed to Xin Wang.

At least 1,477 records · Page 82Linked to original sources

arXiv2020

Rydberg level shift due to the electric field generated by Rydberg atom collision induced ionization in cesium atomic ensemble

We experimentally studied the Rydberg level shift caused by the electric field, which is generated by Rydberg atom collision induced ionization in a cesium atomic ensemble. The density of charged particles caused by collisions between Rydberg atoms is changed by controlling the ground-state atomic density and optical excitation process. We measured the Rydberg level shift using Rydberg electromagnetically-induced-transparency (EIT) spectroscopy, and interpreted the physical origin using a semi-classical model. The experimental results are in good agreement with the numerical simulation. These energy shifts are important for the self-calibrated sensing of microwave field by the employing of Rydberg EIT. Moreover, in contrast to the resonant excitation case, narrow-linewidth spectroscopy with high signal-to-noise ratio would be useful for high-precision measurements.

physics.atom-ph↗

arXiv2020

Joint Modeling of Chest Radiographs and Radiology Reports for Pulmonary Edema Assessment

We propose and demonstrate a novel machine learning algorithm that assesses pulmonary edema severity from chest radiographs. While large publicly available datasets of chest radiographs and free-text radiology reports exist, only limited numerical edema severity labels can be extracted from radiology reports. This is a significant challenge in learning such models for image classification. To take advantage of the rich information present in the radiology reports, we develop a neural network model that is trained on both images and free-text to assess pulmonary edema severity from chest radiographs at inference time. Our experimental results suggest that the joint image-text representation learning improves the performance of pulmonary edema assessment compared to a supervised model trained on images only. We also show the use of the text for explaining the image classification by the joint model. To the best of our knowledge, our approach is the first to leverage free-text radiology reports for improving the image model performance in this application. Our code is available at https://github.com/RayRuizhiLiao/joint_chestxray.

cs.CV↗

arXiv2020

Learning Tuple Compatibility for Conditional OutfitRecommendation

Outfit recommendation requires the answers of some challenging outfit compatibility questions such as 'Which pair of boots and school bag go well with my jeans and sweater?'. It is more complicated than conventional similarity search, and needs to consider not only visual aesthetics but also the intrinsic fine-grained and multi-category nature of fashion items. Some existing approaches solve the problem through sequential models or learning pair-wise distances between items. However, most of them only consider coarse category information in defining fashion compatibility while neglecting the fine-grained category information often desired in practical applications. To better define the fashion compatibility and more flexibly meet different needs, we propose a novel problem of learning compatibility among multiple tuples (each consisting of an item and category pair), and recommending fashion items following the category choices from customers. Our contributions include: 1) Designing a Mixed Category Attention Net (MCAN) which integrates both fine-grained and coarse category information into recommendation and learns the compatibility among fashion tuples. MCAN can explicitly and effectively generate diverse and controllable recommendations based on need. 2) Contributing a new dataset IQON, which follows eastern culture and can be used to test the generalization of recommendation systems. Our extensive experiments on a reference dataset Polyvore and our dataset IQON demonstrate that our method significantly outperforms state-of-the-art recommendation methods.

cs.CV↗

arXiv2020

Eco-evolutionary dynamics with environmental feedback: cooperation in a changing world

Eco-evolutionary game dynamics which characterizes the mutual interactions and the coupled evolutions of strategies and environments has been of growing interests in very recent years. Since such feedback loops widely exist in a range of coevolutionary systems, such as microbial systems, social-ecological system and psychological-economic system, recent modeling frameworks that unveil the oscillating dynamics of social dilemmas have great potential for practical applications. In this perspective article, we overview the latest progress of evolutionary game theory in this direction. We describe both mathematical methods and interdisciplinary applications across different fields. The ideas worthy of further consideration are discussed in prospects, with the central role of promoting cooperations in a changing world.

physics.soc-ph↗

arXiv2020

Tunable optomechanically induced transparency by controlling the dark-mode effect

We study tunable optomechanically induced transparency by controlling the dark-mode effect induced by two mechanical modes coupled to a common cavity field. This is realized by introducing a phase-dependent phonon-exchange interaction, which is used to form a loop-coupled configuration. Combining this phase-dependent coupling with the optomechanical interactions, the dark-mode effect can be controlled by the quantum interference effect. In particular, the dark-mode effect in this two-mechanical-mode optomechanical system can lead to a double-amplified optomechanically induced transparency (OMIT) window and a higher efficiency of the second-order sideband in comparison with the standard optomechanical system. This is because the effective mechanical decay rate related to the linewidth of the OMIT window becomes a twofold increase in the weak-coupling limit. When the dark-mode effect is broken, controllable double transparency windows appear and the second-order sideband, as well as the light delay or advance, is significantly enhanced. For an N-mechanical-mode optomechanical system, we find that in the presence of the dark-mode effect, the amplification multiple of the linewidth of the OMIT window is nearly proportional to the number of mechanical modes, and that the OMIT with a single window becomes the one with N tunable windows by breaking the dark-mode effect. The study will be useful in optical information storage within a large-frequency bandwidth and multichannel optical communication based on optomechanical systems.

quant-ph↗

arXiv2020

Introducing the VoicePrivacy Initiative

The VoicePrivacy initiative aims to promote the development of privacy preservation tools for speech technology by gathering a new community to define the tasks of interest and the evaluation methodology, and benchmarking solutions through a series of challenges. In this paper, we formulate the voice anonymization task selected for the VoicePrivacy 2020 Challenge and describe the datasets used for system development and evaluation. We also present the attack models and the associated objective and subjective evaluation metrics. We introduce two anonymization baselines and report objective evaluation results.

cs.CL↗

arXiv2020

Deep Visual Odometry with Adaptive Memory

We propose a novel deep visual odometry (VO) method that considers global information by selecting memory and refining poses. Existing learning-based methods take the VO task as a pure tracking problem via recovering camera poses from image snippets, leading to severe error accumulation. Global information is crucial for alleviating accumulated errors. However, it is challenging to effectively preserve such information for end-to-end systems. To deal with this challenge, we design an adaptive memory module, which progressively and adaptively saves the information from local to global in a neural analogue of memory, enabling our system to process long-term dependency. Benefiting from global information in the memory, previous results are further refined by an additional refining module. With the guidance of previous outputs, we adopt a spatial-temporal attention to select features for each view based on the co-visibility in feature domain. Specifically, our architecture consisting of Tracking, Remembering and Refining modules works beyond tracking. Experiments on the KITTI and TUM-RGBD datasets demonstrate that our approach outperforms state-of-the-art methods by large margins and produces competitive results against classic approaches in regular scenes. Moreover, our model achieves outstanding performance in challenging scenarios such as texture-less regions and abrupt motions, where classic algorithms tend to fail.

cs.RO↗

arXiv2020

A SLAM Map Restoration Algorithm Based on Submaps and an Undirected Connected Graph

Many visual simultaneous localization and mapping (SLAM) systems have been shown to be accurate and robust, and have real-time performance capabilities on both indoor and ground datasets. However, these methods can be problematic when dealing with aerial frames captured by a camera mounted on an unmanned aerial vehicle (UAV) because the flight height of the UAV can be difficult to control and is easily affected by the environment.To cope with the case of lost tracking, many visual SLAM systems employ a relocalization strategy. This involves the tracking thread continuing the online working by inspecting the connections between the subsequent new frames and the generated map before the tracking was lost. To solve the missing map problem, which is an issue in many applications , after the tracking is lost, based on monocular visual SLAM, we present a method of reconstructing a complete global map of UAV datasets by sequentially merging the submaps via the corresponding undirected connected graph. Specifically, submaps are repeatedly generated, from the initialization process to the place where the tracking is lost, and a corresponding undirected connected graph is built by considering these submaps as nodes and the common map points within two submaps as edges. The common map points are then determined by the bag-of-words (BoW) method, and the submaps are merged if they are found to be connected with the online map in the undirect connected graph. To demonstrate the performance of the proposed method, we first investigated the performance on a UAV dataset, and the experimental results showed that, in the case of several tracking failures, the integrity of the mapping was significantly better than that of the current mainstream SLAM method.

cs.CV↗

arXiv2020

Cost of quantum entanglement simplified

Quantum entanglement is a key physical resource in quantum information processing that allows for performing basic quantum tasks such as teleportation and quantum key distribution, which are impossible in the classical world. Ever since the rise of quantum information theory, it has been an open problem to quantify entanglement in an information-theoretically meaningful way. In particular, every previously defined entanglement measure bearing a precise information-theoretic meaning is not known to be efficiently computable, or if it is efficiently computable, then it is not known to have a precise information-theoretic meaning. In this Letter, we meet this challenge by introducing an entanglement measure that has a precise information-theoretic meaning as the exact cost required to prepare an entangled state when two distant parties are allowed to perform quantum operations that completely preserve the positivity of the partial transpose. Additionally, this entanglement measure is efficiently computable by means of a semidefinite program, and it bears a number of useful properties such as additivity and faithfulness. Our results bring key insights into the fundamental entanglement structure of arbitrary quantum states, and they can be used directly to assess and quantify the entanglement produced in quantum-physical experiments.

quant-ph↗

arXiv2020

A Census of Sub-kiloparsec Resolution Metallicity Gradients in Star-forming Galaxies at Cosmic Noon from HST Slitless Spectroscopy

We present hitherto the largest sample of gas-phase metallicity radial gradients measured at sub-kiloparsec resolution in star-forming galaxies in the redshift range of $z\in[1.2, 2.3]$. These measurements are enabled by the synergy of slitless spectroscopy from the Hubble Space Telescope near-infrared channels and the lensing magnification from foreground galaxy clusters. Our sample consists of 76 galaxies with stellar mass ranging from 10$^7$ to 10$^{10}$ $M_\odot$, instantaneous star-formation rate in the range of [1, 100] $M_\odot$/yr, and global metallicity [$\frac{1}{12}$, 2] solar. At 2-$σ$ confidence level, 15/76 galaxies in our sample show negative radial gradients, whereas 7/76 show inverted gradients. Combining ours and all other metallicity gradients obtained at similar resolution currently available in the literature, we measure a negative mass dependence of $Δ\log({\rm O/H})/Δr~ [\mathrm{dex~kpc^{-1}}] = \left(-0.020\pm0.007\right) + \left(-0.016\pm0.008\right) \log(M_\ast/10^{9.4} M_\odot)$ with the intrinsic scatter being $σ=0.060\pm0.006$ over four orders of magnitude in stellar mass. Our result is consistent with strong feedback, not secular processes, being the primary governor of the chemo-structural evolution of star-forming galaxies during the disk mass assembly at cosmic noon. We also find that the intrinsic scatter of metallicity gradients increases with decreasing stellar mass and increasing specific star-formation rate. This increase in the intrinsic scatter is likely caused by the combined effect of cold-mode gas accretion and merger-induced starbursts, with the latter more predominant in the dwarf mass regime of $M_\ast\lesssim10^9 M_\odot$.

astro-ph.GA↗

arXiv2020

Stochastic switching in Rydberg atomic ensemble

We demonstrated stochastic switching in a bistable system implemented with Rydberg atomic ensemble. The transition between the two states of the bistable system is driven by intensity noise of the laser beams. Rydberg atomic ensemble accumulates energy in an equilibrium situation and brings the nonlinear system across the threshold, where stochastic switching occurs between the two states. Measurement of Rydberg state's population by means of the ladder-type electromagnetically-induced transparency approaching relevant Rydberg state allows us to investigate the nonlinear behavior in Rydberg atomic ensemble based bistable system experimentally.

physics.atom-ph↗

arXiv2020

ASVspoof 2019: A large-scale public database of synthesized, converted and replayed speech

Automatic speaker verification (ASV) is one of the most natural and convenient means of biometric person recognition. Unfortunately, just like all other biometric systems, ASV is vulnerable to spoofing, also referred to as "presentation attacks." These vulnerabilities are generally unacceptable and call for spoofing countermeasures or "presentation attack detection" systems. In addition to impersonation, ASV systems are vulnerable to replay, speech synthesis, and voice conversion attacks. The ASVspoof 2019 edition is the first to consider all three spoofing attack types within a single challenge. While they originate from the same source database and same underlying protocol, they are explored in two specific use case scenarios. Spoofing attacks within a logical access (LA) scenario are generated with the latest speech synthesis and voice conversion technologies, including state-of-the-art neural acoustic and waveform model techniques. Replay spoofing attacks within a physical access (PA) scenario are generated through carefully controlled simulations that support much more revealing analysis than possible previously. Also new to the 2019 edition is the use of the tandem detection cost function metric, which reflects the impact of spoofing and countermeasures on the reliability of a fixed ASV system. This paper describes the database design, protocol, spoofing attack implementations, and baseline ASV and countermeasure results. It also describes a human assessment on spoofed data in logical access. It was demonstrated that the spoofing data in the ASVspoof 2019 database have varied degrees of perceived quality and similarity to the target speakers, including spoofed data that cannot be differentiated from bona-fide utterances even by human subjects.

eess.AS↗

arXiv2020

Lepton Flavor Mixing and CP Violation in the Minimal Type-(I+II) Seesaw Model with a Modular $A_4$ Symmetry

In this paper, we study the implications of the modular $A^{}_4$ flavor symmetry in constructing a supersymmetric minimal type-(I+II) seesaw model, in which only one right-handed neutrino and two Higgs triplets are introduced to account for the tiny neutrino masses, flavor mixing and CP violation. The right-handed neutrino as well as the Higgs triplets in this model are assigned into the trivial one-dimensional irreducible representation of the modular group $A^{}_{4}$. We show that the individual contributions to the neutrino masses from the right-handed neutrino and the Higgs triplet are comparable. We also find that the neutrino mass matrix can possess an approximate $μ-τ$ reflection symmetry for some specific values of free model parameters. Moreover, our model predicts relatively large masses of three light neutrinos, thus can be easily tested in future neutrino experiments.

hep-ph↗

arXiv2020

The curious case of developmental BERTology: On sparsity, transfer learning, generalization and the brain

In this essay, we explore a point of intersection between deep learning and neuroscience, through the lens of large language models, transfer learning and network compression. Just like perceptual and cognitive neurophysiology has inspired effective deep neural network architectures which in turn make a useful model for understanding the brain, here we explore how biological neural development might inspire efficient and robust optimization procedures which in turn serve as a useful model for the maturation and aging of the brain.

cs.CL↗

arXiv2020

Joint User Identification, Channel Estimation, and Signal Detection for Grant-Free NOMA

For massive machine-type communications, centralized control may incur a prohibitively high overhead. Grant-free non-orthogonal multiple access (NOMA) provides possible solutions, yet poses new challenges for efficient receiver design. In this paper, we develop a joint user identification, channel estimation, and signal detection (JUICESD) algorithm. We divide the whole detection scheme into two modules: slot-wise multi-user detection (SMD) and combined signal and channel estimation (CSCE). SMD is designed to decouple the transmissions of different users by leveraging the approximate message passing (AMP) algorithms, and CSCE is designed to deal with the nonlinear coupling of activity state, channel coefficient and transmit signal of each user separately. To address the problem that the exact calculation of the messages exchanged within CSCE and between the two modules is complicated due to phase ambiguity issues, this paper proposes a rotationally invariant Gaussian mixture (RIGM) model, and develops an efficient JUICESD-RIGM algorithm. JUICESD-RIGM achieves a performance close to JUICESD with a much lower complexity. Capitalizing on the feature of RIGM, we further analyze the performance of JUICESD-RIGM with state evolution techniques. Numerical results demonstrate that the proposed algorithms achieve a significant performance improvement over the existing alternatives, and the derived state evolution method predicts the system performance accurately.

eess.SP↗

arXiv2020

Automated Pavement Crack Segmentation Using U-Net-based Convolutional Neural Network

Automated pavement crack image segmentation is challenging because of inherent irregular patterns, lighting conditions, and noise in images. Conventional approaches require a substantial amount of feature engineering to differentiate crack regions from non-affected regions. In this paper, we propose a deep learning technique based on a convolutional neural network to perform segmentation tasks on pavement crack images. Our approach requires minimal feature engineering compared to other machine learning techniques. We propose a U-Net-based network architecture in which we replace the encoder with a pretrained ResNet-34 neural network. We use a "one-cycle" training schedule based on cyclical learning rates to speed up the convergence. Our method achieves an F1 score of 96% on the CFD dataset and 73% on the Crack500 dataset, outperforming other algorithms tested on these datasets. We perform ablation studies on various techniques that helped us get marginal performance boosts, i.e., the addition of spatial and channel squeeze and excitation (SCSE) modules, training with gradually increasing image sizes, and training various neural network layers with different learning rates.

cs.CV↗

arXiv2020

Category-wise Attack: Transferable Adversarial Examples for Anchor Free Object Detection

Deep neural networks have been demonstrated to be vulnerable to adversarial attacks: subtle perturbations can completely change the classification results. Their vulnerability has led to a surge of research in this direction. However, most works dedicated to attacking anchor-based object detection models. In this work, we aim to present an effective and efficient algorithm to generate adversarial examples to attack anchor-free object models based on two approaches. First, we conduct category-wise instead of instance-wise attacks on the object detectors. Second, we leverage the high-level semantic information to generate the adversarial examples. Surprisingly, the generated adversarial examples it not only able to effectively attack the targeted anchor-free object detector but also to be transferred to attack other object detectors, even anchor-based detectors such as Faster R-CNN.

cs.CV↗

arXiv2020

Domain Embedded Multi-model Generative Adversarial Networks for Image-based Face Inpainting

Prior knowledge of face shape and structure plays an important role in face inpainting. However, traditional face inpainting methods mainly focus on the generated image resolution of the missing portion without consideration of the special particularities of the human face explicitly and generally produce discordant facial parts. To solve this problem, we present a domain embedded multi-model generative adversarial model for inpainting of face images with large cropped regions. We firstly represent only face regions using the latent variable as the domain knowledge and combine it with the non-face parts textures to generate high-quality face images with plausible contents. Two adversarial discriminators are finally used to judge whether the generated distribution is close to the real distribution or not. It can not only synthesize novel image structures but also explicitly utilize the embedded face domain knowledge to generate better predictions with consistency on structures and appearance. Experiments on both CelebA and CelebA-HQ face datasets demonstrate that our proposed approach achieved state-of-the-art performance and generates higher quality inpainting results than existing ones.

cs.CV↗