arXiv subjects

Yang Wang

Publications and source records attributed to Yang Wang.

At least 721 records · Page 40Linked to original sources

arXiv2019

Data-driven Method for 3D Axis-symmetric Object Reconstruction from Single Cone-beam Projection Data

In this paper we consider 3D axis-symmetric (AS) object reconstruction from single cone-beam x-ray projection data. Traditional x-ray CT fails to capture fleeting state of material due to the long time for data acquisition at all angles. Therefore, AS object is devised to investigate the instant deformation of material under pulse change of environment because single projection data is enough to reconstruct its inner structure. Previous reconstruction methods are layer by layer, and ignore the longitudinal tilt of x-ray paths. We propose a regularization method using adaptive tight frame to reconstruct the 3D AS object structure simultaneously. Alternating direction method is adopted to solve the proposed model. More importantly, a numerical algorithm is developed to compute imaging matrix. Experiments on simulation data verify the effectiveness of our method

math.NA↗

arXiv2019

Robust Bang-Off-Bang Low-Thrust Guidance Using Model Predictive Static Programming

Model Predictive Static Programming (MPSP) was always used under the assumption of continuous control, which impedes it for applications with bang-off-bang control directly. In this paper, MPSP is employed for the first time as a guidance scheme for low-thrust transfers with bang-off-bang control where the fuel-optimal trajectory is used as the nominal solution. In our method, dynamical equations in Cartesian coordinates are augmented by the mass costate equation, while the unconstrained velocity costate vector is used as control variable, and is expressed as a combination of Fourier basis functions with corresponding weights. A two-loop MPSP algorithm is designed where the weights and the initial mass costate are updated in the inner loop and continuation is conducted on the outer loop in case of large perturbations. The sensitivity matrix (SM) is recursively calculated using analytical derivatives and SM at switching points is compensated based on calculus of variations. An sample interplanetary CubeSat mission to an asteroid is used as study case to illustrate the effectiveness of the method developed.

math.OC↗

arXiv2019

Electronic Structure of Exfoliated Millimeter-Sized Monolayer WSe2 on Silicon Wafer

The monolayer WSe2 is interesting and important for future application in nanoelectronics, spintronics and valleytronics devices, because it has the largest spin splitting and longest valley coherence time among all the known monolayer transition-metal dichalcogenides (TMDs). To obtain the large-area monolayer TMDs' crystal is the first step to manufacture scalable and high-performance electronic devices. In this letter, we have successfully fabricated millimeter-sized monolayer WSe2 single crystals with very high quality, based on our improved mechanical exfoliation method. With such superior samples, using standard high resolution angle-resolved photoemission spectroscopy, we did comprehensive electronic band structure measurements on our monolayer WSe2. The overall band features point it to be a 1.2eV direct band gap semiconductor. Its spin-splitting of the valence band at K point is found as 460 meV, which is 30 meV less than the corresponding band splitting in its bulk counterpart. The effective hole masses of valence bands are determined as 2.344 me at Gamma, and 0.529 me as well as 0.532 me at K for the upper and lower branch of splitting bands, respectively. And screening effect from substrate is shown to substantially impact on the electronic properties. Our results provide important insights into band structure engineering in monolayer TMDs. Our monolayer WSe2 crystals may constitute a valuable device platform.

cond-mat.mes-hall↗

arXiv2019

Learning to Recommend via Meta Parameter Partition

In this paper we propose to solve an important problem in recommendation -- user cold start, based on meta leaning method. Previous meta learning approaches finetune all parameters for each new user, which is both computing and storage expensive. In contrast, we divide model parameters into fixed and adaptive parts and develop a two-stage meta learning algorithm to learn them separately. The fixed part, capturing user invariant features, is shared by all users and is learned during offline meta learning stage. The adaptive part, capturing user specific features, is learned during online meta learning stage. By decoupling user invariant parameters from user dependent parameters, the proposed approach is more efficient and storage cheaper than previous methods. It also has potential to deal with catastrophic forgetting while continually adapting for streaming coming users. Experiments on production data demonstrates that the proposed method converges faster and to a better performance than baseline methods. Meta-training without online meta model finetuning increases the AUC from 72.24% to 74.72% (2.48% absolute improvement). Online meta training achieves a further gain of 2.46\% absolute improvement comparing with offline meta training.

cs.IR↗

arXiv2019

Quantum search for unknown number of target items hybridizing the fixed-point method with the trail-and-error method

For the unsorted database quantum search with the unknown fraction $λ$ of target items, there are mainly two kinds of methods, i.e., fixed-point or trail-and-error. (i) In terms of the fixed-point method, Yoder et al. [Phys. Rev. Lett. 113, 210501 (2014)] claimed that the quadratic speedup over classical algorithms has been achieved. However, in this paper, we point out that this is not the case, because the query complexity of Yoder's algorithm is actually in $O(1/\sqrt{λ_0})$ rather than $O(1/\sqrtλ)$, where $λ_0$ is a known lower bound of $λ$. (ii) In terms of the trail-and-error method, currently the algorithm without randomness has to take more than 1 times queries or iterations than the algorithm with randomly selected parameters. For the above problems, we provide the first hybrid quantum search algorithm based on the fixed-point and trail-and-error methods, where the matched multiphase Grover operations are trialed multiple times and the number of iterations increases exponentially along with the number of trials. The upper bound of expected queries as well as the optimal parameters are derived. Compared with Yoder's algorithm, the query complexity of our algorithm indeed achieves the optimal scaling in $λ$ for quantum search, which reconfirms the practicality of the fixed-point method. In addition, our algorithm also does not contain randomness, and compared with the existing deterministic algorithm, the query complexity can be reduced by about 1/3. Our work provides an new idea for the research on fixed-point and trial-and-error quantum search.

quant-ph↗

arXiv2019

Diversifying Inference Path Selection: Moving-Mobile-Network for Landmark Recognition

Deep convolutional neural networks have largely benefited computer vision tasks. However, the high computational complexity limits their real-world applications. To this end, many methods have been proposed for efficient network learning, and applications in portable mobile devices. In this paper, we propose a novel \underline{M}oving-\underline{M}obile-\underline{Net}work, named M$^2$Net, for landmark recognition, equipped each landmark image with located geographic information. We intuitively find that M$^2$Net can essentially promote the diversity of the inference path (selected blocks subset) selection, so as to enhance the recognition accuracy. The above intuition is achieved by our proposed reward function with the input of geo-location and landmarks. We also find that the performance of other portable networks can be improved via our architecture. We construct two landmark image datasets, with each landmark associated with geographic information, over which we conduct extensive experiments to demonstrate that M$^2$Net achieves improved recognition accuracy with comparable complexity.

cs.CV↗

arXiv2019

Progressive Retinex: Mutually Reinforced Illumination-Noise Perception Network for Low Light Image Enhancement

Contrast enhancement and noise removal are coupled problems for low-light image enhancement. The existing Retinex based methods do not take the coupling relation into consideration, resulting in under or over-smoothing of the enhanced images. To address this issue, this paper presents a novel progressive Retinex framework, in which illumination and noise of low-light image are perceived in a mutually reinforced manner, leading to noise reduction low-light enhancement results. Specifically, two fully pointwise convolutional neural networks are devised to model the statistical regularities of ambient light and image noise respectively, and to leverage them as constraints to facilitate the mutual learning process. The proposed method not only suppresses the interference caused by the ambiguity between tiny textures and image noises, but also greatly improves the computational efficiency. Moreover, to solve the problem of insufficient training data, we propose an image synthesis strategy based on camera imaging model, which generates color images corrupted by illumination-dependent noises. Experimental results on both synthetic and real low-light images demonstrate the superiority of our proposed approaches against the State-Of-The-Art (SOTA) low-light enhancement methods.

eess.IV↗

arXiv2019

Phase retrieval for sub-Gaussian measurements

Generally, phase retrieval problem can be viewed as the reconstruction of a function/signal from only the magnitude of the linear measurements. These measurements can be, for example, the Fourier transform of the density function. Computationally the phase retrieval problem is very challenging. Many algorithms for phase retrieval are based on i.i.d. Gaussian random measurements. However, Gaussian random measurements remain one of the very few classes of measurements. In this paper, we develop an efficient phase retrieval algorithm for sub-gaussian random frames. We provide a general condition for measurements and develop a modified spectral initialization. In the algorithm, we first obtain a good approximation of the solution through the initialization, and from there we useWirtinger Flow to solve for the solution. We prove that the algorithm converges to the global minimizer linearly.

math.OC↗

arXiv2019

Evidence for an Additional Symmetry Breaking from Direct Observation of Band Splitting in the Nematic State of FeSe Superconductor

The iron-based superconductor FeSe has attracted much recent attention because of its simple crystal structure, distinct electronic structure and rich physics exhibited by itself and its derivatives. Determination of its intrinsic electronic structure is crucial to understand its physical properties and superconductivity mechanism. Both theoretical and experimental studies so far have provided a picture that FeSe consists of one hole-like Fermi surface around the Brillouin zone center in its nematic state. Here we report direct observation of two hole-like Fermi surface sheets around the Brillouin zone center, and the splitting of the associated bands, in the nematic state of FeSe by taking high resolution laser-based angle-resolved photoemission measurements. These results indicate that, in addition to nematic order and spin-orbit coupling, there is an additional order in FeSe that breaks either inversion or time reversal symmetries. The new Fermi surface topology asks for reexamination of the existing theoretical and experimental understanding of FeSe and stimulates further efforts to identify the origin of the hidden order in its nematic state.

cond-mat.supr-con↗

arXiv2019

When AWGN-based Denoiser Meets Real Noises

Discriminative learning-based image denoisers have achieved promising performance on synthetic noises such as Additive White Gaussian Noise (AWGN). The synthetic noises adopted in most previous work are pixel-independent, but real noises are mostly spatially/channel-correlated and spatially/channel-variant. This domain gap yields unsatisfied performance on images with real noises if the model is only trained with AWGN. In this paper, we propose a novel approach to boost the performance of a real image denoiser which is trained only with synthetic pixel-independent noise data dominated by AWGN. First, we train a deep model that consists of a noise estimator and a denoiser with mixed AWGN and Random Value Impulse Noise (RVIN). We then investigate Pixel-shuffle Down-sampling (PD) strategy to adapt the trained model to real noises. Extensive experiments demonstrate the effectiveness and generalization of the proposed approach. Notably, our method achieves state-of-the-art performance on real sRGB images in the DND benchmark among models trained with synthetic noises. Codes are available at https://github.com/yzhouas/PD-Denoising-pytorch.

cs.CV↗

arXiv2019

Optimal low rank tensor recovery

We investigate the sample size requirement for exact recovery of a high order tensor of low rank from a subset of its entries. In the Tucker decomposition framework, we show that the Riemannian optimization algorithm with initial value obtained from a spectral method can reconstruct a tensor of size $n\times n \times\cdots \times n$ tensor of ranks $(r,\cdots,r)$ with high probability from as few as $O((r^d+dnr)\log(d))$ entries. In the case of order 3 tensor, the entries can be asymptotically as few as $O(nr)$ for a low rank large tensor. We show the theoretical guarantee condition for the recovery. The analysis relies on the tensor restricted isometry property (tensor RIP) and the curvature of the low rank tensor manifold. Our algorithm is computationally efficient and easy to implement. Numerical results verify that the algorithms are able to recover a low rank tensor from minimum number of measurements. The experiments on hyperspectral images recovery also show that our algorithm is capable of real world signal processing problems.

stat.ML↗

arXiv2019

BigDL: A Distributed Deep Learning Framework for Big Data

This paper presents BigDL (a distributed deep learning framework for Apache Spark), which has been used by a variety of users in the industry for building deep learning applications on production big data platforms. It allows deep learning applications to run on the Apache Hadoop/Spark cluster so as to directly process the production data, and as a part of the end-to-end data analysis pipeline for deployment and management. Unlike existing deep learning frameworks, BigDL implements distributed, data parallel training directly on top of the functional compute model (with copy-on-write and coarse-grained operations) of Spark. We also share real-world experience and "war stories" of users that have adopted BigDL to address their challenges(i.e., how to easily build end-to-end data analysis and deep learning pipelines for their production data).

cs.DC↗

arXiv2019

Gene-Patterns: Should Architecture be Customized for Each Application?

Providing architectural support is crucial for newly arising applications to achieve high performance and high system efficiency. Currently there is a trend in designing accelerators for special applications, while arguably a debate is sparked whether we should customize architecture for each application. In this study, we introduce what we refer to as Gene-Patterns, which are the base patterns of diverse applications. We present a Recursive Reduce methodology to identify the hotspots, and a HOtspot Trace Suite (HOTS) is provided for the research community. We first extract the hotspot patterns, and then, remove the redundancy to obtain the base patterns. We find that although the number of applications is huge and ever-increasing, the amount of base patterns is relatively small, due to the similarity among the patterns of diverse applications. The similarity stems not only from the algorithms but also from the data structures. We build the Periodic Table of Memory Access Patterns (PT-MAP), where the indifference curves are analogous to the energy levels in physics, and memory performance optimization is essentially an energy level transition. We find that inefficiency results from the mismatch between some of the base patterns and the micro-architecture of modern processors. We have identified the key micro-architecture demands of the base patterns. The Gene-Pattern concept, methodology, and toolkit will facilitate the design of both hardware and software for the matching between architectures and applications.

cs.DC↗

arXiv2019

Efficient EM-Variational Inference for Hawkes Process

In classical Hawkes process, the baseline intensity and triggering kernel are assumed to be a constant and parametric function respectively, which limits the model flexibility. To generalize it, we present a fully Bayesian nonparametric model, namely Gaussian process modulated Hawkes process and propose an EM-variational inference scheme. In this model, a transformation of Gaussian process is used as a prior on the baseline intensity and triggering kernel. By introducing a latent branching structure, the inference of baseline intensity and triggering kernel is decoupled and the variational inference scheme is embedded into an EM framework naturally. We also provide a series of schemes to accelerate the inference. Results of synthetic and real data experiments show that the underlying baseline intensity and triggering kernel can be recovered without parametric restriction and our Bayesian nonparametric estimation is superior to other state of the arts.

cs.LG↗

arXiv2019

Scalable Inference for Nonparametric Hawkes Process Using Pólya-Gamma Augmentation

In this paper, we consider the sigmoid Gaussian Hawkes process model: the baseline intensity and triggering kernel of Hawkes process are both modeled as the sigmoid transformation of random trajectories drawn from Gaussian processes (GP). By introducing auxiliary latent random variables (branching structure, Pólya-Gamma random variables and latent marked Poisson processes), the likelihood is converted to two decoupled components with a Gaussian form which allows for an efficient conjugate analytical inference. Using the augmented likelihood, we derive an expectation-maximization (EM) algorithm to obtain the maximum a posteriori (MAP) estimate. Furthermore, we extend the EM algorithm to an efficient approximate Bayesian inference algorithm: mean-field variational inference. We demonstrate the performance of two algorithms on simulated fictitious data. Experiments on real data show that our proposed inference algorithms can recover well the underlying prompting characteristics efficiently.

cs.LG↗

arXiv2019

Finite-key analysis for a practical decoy-state twin-field quantum key distribution

Twin-field quantum key distribution (TF-QKD), which is immune to all possible detector side channel attacks, enables two remote legitimate users to perform secure communications without quantum repeaters. With the help of a central node, TF-QKD is expected to overcome the linear key-rate constraint using current technologies. However, the security of the former TF-QKD protocols relies on the hypothesis of infinite-key and stable sources. In this paper, we present the finite-key analysis of a practical decoy-state twin-field quantum key distribution with variant statistical fluctuation models. We examine the composable security of the protocol with intensity fluctuations of unstable sources employing Azuma's inequality. Our simulation results indicate that the secret key rate is able to surpass the linear key-rate bound with limited signal pulses and intensity fluctuations. In addition, the effect of intensity fluctuations is extremely significant for small size of total signals.

quant-ph↗

arXiv2019

Region Mutual Information Loss for Semantic Segmentation

Semantic segmentation is a fundamental problem in computer vision. It is considered as a pixel-wise classification problem in practice, and most segmentation models use a pixel-wise loss as their optimization riterion. However, the pixel-wise loss ignores the dependencies between pixels in an image. Several ways to exploit the relationship between pixels have been investigated, \eg, conditional random fields (CRF) and pixel affinity based methods. Nevertheless, these methods usually require additional model branches, large extra memories, or more inference time. In this paper, we develop a region mutual information (RMI) loss to model the dependencies among pixels more simply and efficiently. In contrast to the pixel-wise loss which treats the pixels as independent samples, RMI uses one pixel and its neighbour pixels to represent this pixel. Then for each pixel in an image, we get a multi-dimensional point that encodes the relationship between pixels, and the image is cast into a multi-dimensional distribution of these high-dimensional points. The prediction and ground truth thus can achieve high order consistency through maximizing the mutual information (MI) between their multi-dimensional distributions. Moreover, as the actual value of the MI is hard to calculate, we derive a lower bound of the MI and maximize the lower bound to maximize the real value of the MI. RMI only requires a few extra computational resources in the training stage, and there is no overhead during testing. Experimental results demonstrate that RMI can achieve substantial and consistent improvements in performance on PASCAL VOC 2012 and CamVid datasets. The code is available at https://github.com/ZJULearning/RMI.

cs.CV↗

arXiv2019

Deck.gl: Large-scale Web-based Visual Analytics Made Easy

In this paper, we demonstrate how deck.gl, an open-source project born out of data-heavy visual analytics applications, has grown into the robust visualization framework it is today. We begin by explaining why we built another data visualization framework in the first place. Then, we summarize our design goals (distilled from our interactions with users) and discuss how they guided the development of the framework's main features. We use two real-world applications of deck.gl to showcase how it can be applied to simplify the creation of data-heavy visualizations. We also discuss our lessons learned as we continue to improve the framework for the larger visualization community.

cs.HC↗