arXiv subjects

Xiao Wang

Publications and source records attributed to Xiao Wang.

At least 613 records · Page 34Linked to original sources

arXiv2021

Guidance and Teaching Network for Video Salient Object Detection

Owing to the difficulties of mining spatial-temporal cues, the existing approaches for video salient object detection (VSOD) are limited in understanding complex and noisy scenarios, and often fail in inferring prominent objects. To alleviate such shortcomings, we propose a simple yet efficient architecture, termed Guidance and Teaching Network (GTNet), to independently distil effective spatial and temporal cues with implicit guidance and explicit teaching at feature- and decision-level, respectively. To be specific, we (a) introduce a temporal modulator to implicitly bridge features from motion into the appearance branch, which is capable of fusing cross-modal features collaboratively, and (b) utilise motion-guided mask to propagate the explicit cues during the feature aggregation. This novel learning strategy achieves satisfactory results via decoupling the complex spatial-temporal cues and mapping informative cues across different modalities. Extensive experiments on three challenging benchmarks show that the proposed method can run at ~28 fps on a single TITAN Xp GPU and perform competitively against 14 cutting-edge baselines.

cs.CV↗

arXiv2021

Polar or nonpolar? That is not the question for perovskite solar cells

Perovskite solar cell (PSC) is one of the most promising next generation photovoltaic technologies, and there are considerable interests in the role of possible polarization of organic-inorganic halide perovskites (OIHPs) in photovoltaic conversion. The polarity of OIHPs, however, is still hotly debated. In this review, we examine recent literature on the polarity of OIHPs from both theoretical and experimental points of view, and argue that they can be both polar and nonpolar, depending on compositions, processing, and environments. Implications of OIHP polarity to photovoltaic conversion is also discussed, and effort in answering these questions continues to render us new insights. In the future, integrating local scanning probe with global macroscopic measurements in-situ will provide invaluable microscopic insight into the intriguing macroscopic phenomena, while synchrotron diffractions and scanning transmission electron microscopy on more stable samples may ultimately settle the debate.

physics.app-ph↗

arXiv2021

Self-supervised Heterogeneous Graph Neural Network with Co-contrastive Learning

Heterogeneous graph neural networks (HGNNs) as an emerging technique have shown superior capacity of dealing with heterogeneous information network (HIN). However, most HGNNs follow a semi-supervised learning manner, which notably limits their wide use in reality since labels are usually scarce in real applications. Recently, contrastive learning, a self-supervised method, becomes one of the most exciting learning paradigms and shows great potential when there are no labels. In this paper, we study the problem of self-supervised HGNNs and propose a novel co-contrastive learning mechanism for HGNNs, named HeCo. Different from traditional contrastive learning which only focuses on contrasting positive and negative samples, HeCo employs cross-viewcontrastive mechanism. Specifically, two views of a HIN (network schema and meta-path views) are proposed to learn node embeddings, so as to capture both of local and high-order structures simultaneously. Then the cross-view contrastive learning, as well as a view mask mechanism, is proposed, which is able to extract the positive and negative embeddings from two views. This enables the two views to collaboratively supervise each other and finally learn high-level node embeddings. Moreover, two extensions of HeCo are designed to generate harder negative samples with high quality, which further boosts the performance of HeCo. Extensive experiments conducted on a variety of real-world networks show the superior performance of the proposed methods over the state-of-the-arts.

cs.LG↗

arXiv2021

Energy budget and the gravitational wave spectra beyond the bag model

The energy budget of cosmological first-order phase transition is essential for the gravitational wave spectra. Most of the previous studies are based on the bag model with same sound velocity in the symmetric and broken phase. We study the energy budget and the corresponding gravitational wave spectra beyond the bag model, where the sound velocities could be different in the symmetric and broken phase. Taking the Higgs sextic effective model as a representative model, we calculate the sound velocities in different phase, the gravitational wave spectra and the signal-to-noise ratio for different combinations of phase transition parameters beyond the bag model. We compare these new results with the ones obtained from the bag model.The proper sound velocities and phase transition parameters at the appropriate temperature are important to obtain more precise predictions.

astro-ph.CO↗

arXiv2021

Substitutional Neural Image Compression

We describe Substitutional Neural Image Compression (SNIC), a general approach for enhancing any neural image compression model, that requires no data or additional tuning of the trained model. It boosts compression performance toward a flexible distortion metric and enables bit-rate control using a single model instance. The key idea is to replace the image to be compressed with a substitutional one that outperforms the original one in a desired way. Finding such a substitute is inherently difficult for conventional codecs, yet surprisingly favorable for neural compression models thanks to their fully differentiable structures. With gradients of a particular loss backpropogated to the input, a desired substitute can be efficiently crafted iteratively. We demonstrate the effectiveness of SNIC, when combined with various neural compression models and target metrics, in improving compression quality and performing bit-rate control measured by rate-distortion curves. Empirical results of control precision and generation speed are also discussed.

cs.CV↗

arXiv2021

High-Robustness, Low-Transferability Fingerprinting of Neural Networks

This paper proposes Characteristic Examples for effectively fingerprinting deep neural networks, featuring high-robustness to the base model against model pruning as well as low-transferability to unassociated models. This is the first work taking both robustness and transferability into consideration for generating realistic fingerprints, whereas current methods lack practical assumptions and may incur large false positive rates. To achieve better trade-off between robustness and transferability, we propose three kinds of characteristic examples: vanilla C-examples, RC-examples, and LTRC-example, to derive fingerprints from the original base model. To fairly characterize the trade-off between robustness and transferability, we propose Uniqueness Score, a comprehensive metric that measures the difference between robustness and transferability, which also serves as an indicator to the false alarm problem.

cs.LG↗

arXiv2021

Dynamical behavior of a nonlocal Fokker-Planck equation for a stochastic system with tempered stable noise

We characterize a stochastic dynamical system with tempered stable noise, by examining its probability density evolution. This probability density function satisfies a nonlocal Fokker-Planck equation. First, we prove a superposition principle that the probability measure-valued solution to this nonlocal Fokker-Planck equation is equivalent to the martingale solution composed with the inverse stochastic flow. This result together with a Schauder estimate leads to the existence and uniqueness of strong solution for the nonlocal Fokker-Planck equation. Second, we devise a convergent finite difference method to simulate the probability density function by solving the nonlocal Fokker-Planck equation. Finally, we apply our aforementioned theoretical and numerical results to a nonlinear filtering system by simulating a nonlocal Zakai equation.

math.DS↗

arXiv2021

Persistent Ionic Photo-responses and Frank-Condon Mechanism in Proton-transfer Ferroelectrics

Photoexcitation is well-known to trigger electronic metastable states and lead to phenomena like long-lived photoluminescence and photoconductivity. In contrast, persistent photo-response due to ionic metastable states is rare. In this work, we report persistent structural and ferroelectric photo-responses due to proton metastable states via a nuclear quantum mechanism in ferroelectric croconic acid, in which the proton-transfer origin of ferroelectricity is important for the ionic metastable states. We show that, after photoexcitation, the changes of structural and ferroelectric properties relax in about 1000 s, while the photoconductivity decays within 1 s, indicating the dominant ionic origin of the responses. The photogenerated internal bias field that survives polarization switching process suggests another proton transfer route and metastable state, in addition to the metastable states resulting from proton transfer along the hydrogen bonds proposed previously. Analysis based on Frank Condon principle reveals the quantum mechanical nature of the proton-transfer process both within the hydrogen bonds and out of the hydrogen bonds, where small mass of proton and significant change of potential landscape due to the excited electronic states are the key. The demonstration of persistent photo-responses due to the proton metastable states unveils a nuclear quantum mechanism for photo-tunability of materials, which is expected to impact many material properties sensitive to ionic positions.

cond-mat.mtrl-sci↗

arXiv2021

TextFlint: Unified Multilingual Robustness Evaluation Toolkit for Natural Language Processing

Various robustness evaluation methodologies from different perspectives have been proposed for different natural language processing (NLP) tasks. These methods have often focused on either universal or task-specific generalization capabilities. In this work, we propose a multilingual robustness evaluation platform for NLP tasks (TextFlint) that incorporates universal text transformation, task-specific transformation, adversarial attack, subpopulation, and their combinations to provide comprehensive robustness analysis. TextFlint enables practitioners to automatically evaluate their models from all aspects or to customize their evaluations as desired with just a few lines of code. To guarantee user acceptability, all the text transformations are linguistically based, and we provide a human evaluation for each one. TextFlint generates complete analytical reports as well as targeted augmented data to address the shortcomings of the model's robustness. To validate TextFlint's utility, we performed large-scale empirical evaluations (over 67,000 evaluations) on state-of-the-art deep learning models, classic supervised methods, and real-world systems. Almost all models showed significant performance degradation, including a decline of more than 50% of BERT's prediction accuracy on tasks such as aspect-level sentiment classification, named entity recognition, and natural language inference. Therefore, we call for the robustness to be included in the model evaluation, so as to promote the healthy development of NLP technology.

cs.CL↗

arXiv2021

Stacking faults in $α$-RuCl$_3$ revealed by local electric polarization

We present out-of-plane dielectric and magnetodielectric measurements of single crystallines $α$-RuCl$_3$ with various degrees of stack faults. A frequency dependent, but field independent, dielectric anomaly appears at $T_{A}\:(f=100\:\mathrm{kHz})\sim$ 4 K once both magnetic transitions at $T_{N1}\sim$ 7 K and $T_{N2}\sim$ 14 K set in. The observed dielectric anomaly is attributed to the emergency of possible local electric polarizations whose inversion symmetry is broken by inhomogeneously distributed stacking faults. A field-induced intermediate phase is only observed when a magnetic field is applied perpendicular to the Ru-Ru bonds for samples with minimal stacking faults. Less pronounced in-plane anisotropy is found in samples with sizable contribution from stacking imperfections. Our findings suggest that dielectric measurement is a sensitive probe in detecting the structural and magnetic properties, which may be a promising tool especially in studying $α$-RuCl$_3$ thin film devices. Moreover, the stacking details of RuCl$_3$ layers strongly affect the ground state both in the magnetic and electric channels. Such a fragile ground state against stacking faults needs to be overcome for realistic applications utilizing the magnetic and/or electric properties of Kitaev based physics in $α$-RuCl$_3$.

cond-mat.str-el↗

arXiv2021

Performance Portable Back-projection Algorithms on CPUs: Agnostic Data Locality and Vectorization Optimizations

Computed Tomography (CT) is a key 3D imaging technology that fundamentally relies on the compute-intense back-projection operation to generate 3D volumes. GPUs are typically used for back-projection in production CT devices. However, with the rise of power-constrained micro-CT devices, and also the emergence of CPUs comparable in performance to GPUs, back-projection for CPUs could become favorable. Unlike GPUs, extracting parallelism for back-projection algorithms on CPUs is complex given that parallelism and locality are not explicitly defined and controlled by the programmer, as is the case when using CUDA for instance. We propose a collection of novel back-projection algorithms that reduce the arithmetic computation, robustly enable vectorization, enforce a regular memory access pattern, and maximize the data locality. We also implement the novel algorithms as efficient back-projection kernels that are performance portable over a wide range of CPUs. Performance evaluation using a variety of CPUs from different vendors and generations demonstrates that our back-projection implementation achieves on average 5.2x speedup over the multi-threaded implementation of the most widely used, and optimized, open library. With a state-of-the-art CPU, we reach performance that rivals top-performing GPUs.

cs.DC↗

arXiv2021

Physics-Based Iterative Reconstruction for Dual Source and Flying Focal Spot Computed Tomography

For single source helical Computed Tomography (CT), both Filtered-Back Projection (FBP) and statistical iterative reconstruction have been investigated. However for dual source CT with flying focal spot (DS-FFS CT), statistical iterative reconstruction that accurately models the scanner geometry and physics remains unknown to researchers. Therefore, this paper presents a novel physics-based iterative reconstruction method for DS-FFS CT and assess its image quality. Our algorithm uses precise physics models to reconstruct from the native cone-beam geometry and interleaved dual source helical trajectory of a DS-FFS CT. To do so, we construct a noise physics model to represent data acquisition noise and a prior image model to represent image noise and texture. In addition, we design forward system models to compute the locations of deflected focal spots, the dimension and sensitivity of voxels and detector units, as well as the length of intersection between X-rays and voxels. The forward system models further represent the coordinated movement between the dual sources by computing their X-ray coverage gaps and overlaps at an arbitrary helical pitch. With the above models, we reconstruct images by using an advanced Consensus Equilibrium (CE) numerical method to compute the maximum a posteriori estimate to a joint optimization problem that simultaneously fits all models. We compared our reconstruction with Siemens ADMIRE, which is the clinical standard hybrid iterative reconstruction (IR) method for DS-FFS CT, in terms of spatial resolution, noise profile and image artifacts through both phantoms and clinical datasets. Experiments show that our reconstruction has a consistently higher spatial resolution than the clinical standard hybrid IR. In addition, our reconstruction shows a reduced magnitude of image undersampling artifacts than the clinical standard.

eess.IV↗

arXiv2021

Lorentzian Graph Convolutional Networks

Graph convolutional networks (GCNs) have received considerable research attention recently. Most GCNs learn the node representations in Euclidean geometry, but that could have a high distortion in the case of embedding graphs with scale-free or hierarchical structure. Recently, some GCNs are proposed to deal with this problem in non-Euclidean geometry, e.g., hyperbolic geometry. Although hyperbolic GCNs achieve promising performance, existing hyperbolic graph operations actually cannot rigorously follow the hyperbolic geometry, which may limit the ability of hyperbolic geometry and thus hurt the performance of hyperbolic GCNs. In this paper, we propose a novel hyperbolic GCN named Lorentzian graph convolutional network (LGCN), which rigorously guarantees the learned node features follow the hyperbolic geometry. Specifically, we rebuild the graph operations of hyperbolic GCNs with Lorentzian version, e.g., the feature transformation and non-linear activation. Also, an elegant neighborhood aggregation method is designed based on the centroid of Lorentzian distance. Moreover, we prove some proposed graph operations are equivalent in different types of hyperbolic geometry, which fundamentally indicates their correctness. Experiments on six datasets show that LGCN performs better than the state-of-the-art methods. LGCN has lower distortion to learn the representation of tree-likeness graphs compared with existing hyperbolic GCNs. We also find that the performance of some hyperbolic GCNs can be improved by simply replacing the graph operations with those we defined in this paper.

cs.LG↗

arXiv2021

Towards More Flexible and Accurate Object Tracking with Natural Language: Algorithms and Benchmark

Tracking by natural language specification is a new rising research topic that aims at locating the target object in the video sequence based on its language description. Compared with traditional bounding box (BBox) based tracking, this setting guides object tracking with high-level semantic information, addresses the ambiguity of BBox, and links local and global search organically together. Those benefits may bring more flexible, robust and accurate tracking performance in practical scenarios. However, existing natural language initialized trackers are developed and compared on benchmark datasets proposed for tracking-by-BBox, which can't reflect the true power of tracking-by-language. In this work, we propose a new benchmark specifically dedicated to the tracking-by-language, including a large scale dataset, strong and diverse baseline methods. Specifically, we collect 2k video sequences (contains a total of 1,244,340 frames, 663 words) and split 1300/700 for the train/testing respectively. We densely annotate one sentence in English and corresponding bounding boxes of the target object for each video. We also introduce two new challenges into TNL2K for the object tracking task, i.e., adversarial samples and modality switch. A strong baseline method based on an adaptive local-global-search scheme is proposed for future works to compare. We believe this benchmark will greatly boost related researches on natural language guided tracking.

cs.CV↗

arXiv2021

Dynamic Attention guided Multi-Trajectory Analysis for Single Object Tracking

Most of the existing single object trackers track the target in a unitary local search window, making them particularly vulnerable to challenging factors such as heavy occlusions and out-of-view movements. Despite the attempts to further incorporate global search, prevailing mechanisms that cooperate local and global search are relatively static, thus are still sub-optimal for improving tracking performance. By further studying the local and global search results, we raise a question: can we allow more dynamics for cooperating both results? In this paper, we propose to introduce more dynamics by devising a dynamic attention-guided multi-trajectory tracking strategy. In particular, we construct dynamic appearance model that contains multiple target templates, each of which provides its own attention for locating the target in the new frame. Guided by different attention, we maintain diversified tracking results for the target to build multi-trajectory tracking history, allowing more candidates to represent the true target trajectory. After spanning the whole sequence, we introduce a multi-trajectory selection network to find the best trajectory that delivers improved tracking performance. Extensive experimental results show that our proposed tracking strategy achieves compelling performance on various large-scale tracking benchmarks. The project page of this paper can be found at https://sites.google.com/view/mt-track/.

cs.CV↗

arXiv2021

CaPC Learning: Confidential and Private Collaborative Learning

Machine learning benefits from large training datasets, which may not always be possible to collect by any single entity, especially when using privacy-sensitive data. In many contexts, such as healthcare and finance, separate parties may wish to collaborate and learn from each other's data but are prevented from doing so due to privacy regulations. Some regulations prevent explicit sharing of data between parties by joining datasets in a central location (confidentiality). Others also limit implicit sharing of data, e.g., through model predictions (privacy). There is currently no method that enables machine learning in such a setting, where both confidentiality and privacy need to be preserved, to prevent both explicit and implicit sharing of data. Federated learning only provides confidentiality, not privacy, since gradients shared still contain private information. Differentially private learning assumes unreasonably large datasets. Furthermore, both of these learning paradigms produce a central model whose architecture was previously agreed upon by all parties rather than enabling collaborative learning where each party learns and improves their own local model. We introduce Confidential and Private Collaborative (CaPC) learning, the first method provably achieving both confidentiality and privacy in a collaborative setting. We leverage secure multi-party computation (MPC), homomorphic encryption (HE), and other techniques in combination with privately aggregated teacher models. We demonstrate how CaPC allows participants to collaborate without having to explicitly join their training sets or train a central model. Each party is able to improve the accuracy and fairness of their model, even in settings where each party has a model that performs well on their own dataset or when datasets are not IID and model architectures are heterogeneous across parties.

cs.LG↗

arXiv2021

A Negotiation-based Right-of-way Assignment Strategy to Ensure Traffic Safety and Efficiency in Lane Change

It is widely acknowledged that verifying the safety of autonomous driving strategies requires a substantial body of simulation testing and road testing. In recent years, the formal safety methods represented by Responsibility-Sensitive Safety (RSS) have encouraged low-cost autonomous driving safety research, benefitting from its accurate assessment of safety and clear division of responsibilities. However, how to maintain traffic efficiency while ensuring safety remains a challenge. To address this problem, this paper proposes a formulized negotiation-based lane-changing strategy that makes a trade-off between safety and efficiency. Both theoretical analysis and numerical experimental results shows that compared to RSS, our strategy can noticeably improve the success rate of changing lanes on the premise of safety.

eess.SY↗

arXiv2021

Laser Compression via fast-extending plasma gratings

It is proposed a new method of compressing laser pulse by fast extending plasma gratings(FEPG), which is created by ionizing the hypersound wave generated by stimulated Brillouin scattering(SBS) in the background gas. Ionized by a short laser pulse, the phonon forms a light-velocity FEPG to fully reflect a resonant pump laser. As the reflecting surface moves with a light velocity, the reflected pulse is temporally overlapped and compressed. This regime is supported by the simulation results of a fully kinetic particle-in-cell(PIC) code Opic with a laser wavelength of 1um, displaying a pump pulse is compressed from 13ps to a few cycles(7.2fs), with an efficiency close to 80%. It is a promising method to produce critical laser powers due to several features: high efficiency without a linear stage, robustness to plasma instabilities, no seed and a wide range of pump intensity.

physics.plasm-ph↗