arXiv subjects

Cong Xu

Publications and source records attributed to Cong Xu.

At least 73 records · Page 4Linked to original sources

arXiv2022

Tighter uncertainty relations based on $(α,β,γ)$ modified weighted Wigner-Yanase-Dyson skew information of quantum channels

We use a novel formation to illustrate the ($α,β,γ$) modified weighted Wigner-Yanase-Dyson (($α,β,γ$) MWWYD) skew information of quantum channels. By using operator norm inequalities, we explore the sum uncertainty relations for arbitrary $N$ quantum channels and for unitary channels. These uncertainty inequalities are shown to be tighter than the existing ones by a detailed example. Our results are also applicable to the modified weighted Wigner-Yanase-Dyson (MWWYD) skew information and the ($α,γ$) modified weighted Wigner-Yanase-Dyson (($α,γ$) MWWYD) skew information of quantum channels as special cases.

quant-ph↗

arXiv2022

Sum uncertainty relations based on $(α,β,γ)$ weighted Wigner-Yanase-Dyson skew information

We introduce ($α,β,γ$) weighted Wigner-Yanase-Dyson (($α,β,γ$) WWYD) skew information and ($α,β,γ$) modified weighted Wigner-Yanase-Dyson (($α,β,γ$) MWWYD) skew information. We explore the sum uncertainty relations for arbitrary $N$ mutually noncommutative observables based on ($α,β,γ$) WWYD skew information. A series of uncertainty inequalities are derived. We show by detailed example that our results cover and improve the previous ones based on the original Wigner-Yanase (WY) skew information. Finally, we establish new sum uncertainty relations in terms of the ($α,β,γ$) MWWYD skew information for arbitrary $N$ quantum channels.

quant-ph↗

arXiv2022

Uncertainty of quantum channels via modified generalized variance and modified generalized Wigner-Yanase-Dyson skew information

Uncertainty relation is a fundamental issue in quantum mechanics and quantum information theory. By using modified generalized variance (MGV), and modified generalized Wigner-Yanase-Dyson skew information (MGWYD), we identify the total and quantum uncertainty of quantum channels. The elegant properties of the total uncertainty of quantum channels are explored in detail. In addition, we present a trade-off relation between the total uncertainty of quantum channels and the entanglement fidelity and establish the relationships between the total uncertainty and entropy exchange/coherent information. Detailed examples are given to the explicit formulas of the total uncertainty and the quantum uncertainty of quantum channels. Moreover, utilizing a realizable experimental measurement scheme by using the Mach-Zehnder interferometer proposed in Nirala et al. (Phys Rev A 99:022111, 2019), we discuss how to measure the total/quantum uncertainty of quantum channels for pure states.

quant-ph↗

arXiv2022

Sparse-based Domain Adaptation Network for OCTA Image Super-Resolution Reconstruction

Retinal Optical Coherence Tomography Angiography (OCTA) with high-resolution is important for the quantification and analysis of retinal vasculature. However, the resolution of OCTA images is inversely proportional to the field of view at the same sampling frequency, which is not conducive to clinicians for analyzing larger vascular areas. In this paper, we propose a novel Sparse-based domain Adaptation Super-Resolution network (SASR) for the reconstruction of realistic 6x6 mm2/low-resolution (LR) OCTA images to high-resolution (HR) representations. To be more specific, we first perform a simple degradation of the 3x3 mm2/high-resolution (HR) image to obtain the synthetic LR image. An efficient registration method is then employed to register the synthetic LR with its corresponding 3x3 mm2 image region within the 6x6 mm2 image to obtain the cropped realistic LR image. We then propose a multi-level super-resolution model for the fully-supervised reconstruction of the synthetic data, guiding the reconstruction of the realistic LR images through a generative-adversarial strategy that allows the synthetic and realistic LR images to be unified in the feature domain. Finally, a novel sparse edge-aware loss is designed to dynamically optimize the vessel edge structure. Extensive experiments on two OCTA sets have shown that our method performs better than state-of-the-art super-resolution reconstruction methods. In addition, we have investigated the performance of the reconstruction results on retina structure segmentations, which further validate the effectiveness of our approach.

eess.IV↗

arXiv2022

Adversarial Momentum-Contrastive Pre-Training

Recently proposed adversarial self-supervised learning methods usually require big batches and long training epochs to extract robust features, which will bring heavy computational overhead on platforms with limited resources. In order to help the network learn more powerful feature representations in smaller batches and fewer epochs, this paper proposes a novel adversarial momentum contrastive learning method, which introduces two memory banks corresponding to clean samples and adversarial samples, respectively. These memory banks can be dynamically incorporated into the training process to track invariant features among historical mini-batches. Compared with the previous adversarial pre-training model, our method achieves superior performance with smaller batch size and less training epochs. In addition, the model outperforms some state-of-the-art supervised defensive methods on multiple benchmark datasets after being fine-tuned on downstream classification tasks.

cs.CV↗

arXiv2022

Improve Deep Image Inpainting by Emphasizing the Complexity of Missing Regions

Deep image inpainting research mainly focuses on constructing various neural network architectures or imposing novel optimization objectives. However, on the one hand, building a state-of-the-art deep inpainting model is an extremely complex task, and on the other hand, the resulting performance gains are sometimes very limited. We believe that besides the frameworks of inpainting models, lightweight traditional image processing techniques, which are often overlooked, can actually be helpful to these deep models. In this paper, we enhance the deep image inpainting models with the help of classical image complexity metrics. A knowledge-assisted index composed of missingness complexity and forward loss is presented to guide the batch selection in the training procedure. This index helps find samples that are more conducive to optimization in each iteration and ultimately boost the overall inpainting performance. The proposed approach is simple and can be plugged into many deep inpainting models by changing only a few lines of code. We experimentally demonstrate the improvements for several recently developed image inpainting models on various datasets.

cs.CV↗

arXiv2021

An Orthogonal Classifier for Improving the Adversarial Robustness of Neural Networks

Neural networks are susceptible to artificially designed adversarial perturbations. Recent efforts have shown that imposing certain modifications on classification layer can improve the robustness of the neural networks. In this paper, we explicitly construct a dense orthogonal weight matrix whose entries have the same magnitude, thereby leading to a novel robust classifier. The proposed classifier avoids the undesired structural redundancy issue in previous work. Applying this classifier in standard training on clean data is sufficient to ensure the high accuracy and good robustness of the model. Moreover, when extra adversarial samples are used, better robustness can be further obtained with the help of a special worst-case loss. Experimental results show that our method is efficient and competitive to many state-of-the-art defensive approaches. Our code is available at \url{https://github.com/MTandHJ/roboc}.

cs.CV↗

arXiv2021

Change point detection and image segmentation for time series of astrophysical images

Many astrophysical phenomena are time-varying, in the sense that their intensity, energy spectrum, and/or the spatial distribution of the emission suddenly change. This paper develops a method for modeling a time series of images. Under the assumption that the arrival times of the photons follow a Poisson process, the data are binned into 4D grids of voxels (time, energy band, and x-y coordinates), and viewed as a time series of non-homogeneous Poisson images. The method assumes that at each time point, the corresponding multi-band image stack is an unknown 3D piecewise constant function including Poisson noise. It also assumes that all image stacks between any two adjacent change points (in time domain) share the same unknown piecewise constant function. The proposed method is designed to estimate the number and the locations of all the change points (in time domain), as well as all the unknown piecewise constant functions between any pairs of the change points. The method applies the minimum description length (MDL) principle to perform this task. A practical algorithm is also developed to solve the corresponding complicated optimization problem. Simulation experiments and applications to real datasets show that the proposed method enjoys very promising empirical properties. Applications to two real datasets, the XMM observation of a flaring star and an emerging solar coronal loop, illustrate the usage of the proposed method and the scientific insight gained from it.

astro-ph.IM↗

arXiv2020

Improve Adversarial Robustness via Weight Penalization on Classification Layer

It is well-known that deep neural networks are vulnerable to adversarial attacks. Recent studies show that well-designed classification parts can lead to better robustness. However, there is still much space for improvement along this line. In this paper, we first prove that, from a geometric point of view, the robustness of a neural network is equivalent to some angular margin condition of the classifier weights. We then explain why ReLU type function is not a good choice for activation under this framework. These findings reveal the limitations of the existing approaches and lead us to develop a novel light-weight-penalized defensive method, which is simple and has a good scalability. Empirical results on multiple benchmark datasets demonstrate that our method can effectively improve the robustness of the network without requiring too much additional computation, while maintaining a high classification precision for clean data.

cs.LG↗

arXiv2019

A Fast deflation Method for Sparse Principal Component Analysis via Subspace Projections

The implementation of conventional sparse principal component analysis (SPCA) on high-dimensional data sets has become a time consuming work. In this paper, a series of subspace projections are constructed efficiently by using Household QR factorization. With the aid of these subspace projections, a fast deflation method, called SPCA-SP, is developed for SPCA. This method keeps a good tradeoff between various criteria, including sparsity, orthogonality, explained variance, balance of sparsity, and computational cost. Comparative experiments on the benchmark data sets confirm the effectiveness of the proposed method.

stat.ML↗

arXiv2018

SmoothOut: Smoothing Out Sharp Minima to Improve Generalization in Deep Learning

In Deep Learning, Stochastic Gradient Descent (SGD) is usually selected as a training method because of its efficiency; however, recently, a problem in SGD gains research interest: sharp minima in Deep Neural Networks (DNNs) have poor generalization; especially, large-batch SGD tends to converge to sharp minima. It becomes an open question whether escaping sharp minima can improve the generalization. To answer this question, we propose SmoothOut framework to smooth out sharp minima in DNNs and thereby improve generalization. In a nutshell, SmoothOut perturbs multiple copies of the DNN by noise injection and averages these copies. Injecting noises to SGD is widely used in the literature, but SmoothOut differs in lots of ways: (1) a de-noising process is applied before parameter updating; (2) noise strength is adapted to filter norm; (3) an alternative interpretation on the advantage of noise injection, from the perspective of sharpness and generalization; (4) usage of uniform noise instead of Gaussian noise. We prove that SmoothOut can eliminate sharp minima. Training multiple DNN copies is inefficient, we further propose an unbiased stochastic SmoothOut which only introduces the overhead of noise injecting and de-noising per batch. An adaptive variant of SmoothOut, AdaSmoothOut, is also proposed to improve generalization. In a variety of experiments, SmoothOut and AdaSmoothOut consistently improve generalization in both small-batch and large-batch training on the top of state-of-the-art solutions.

cs.LG↗

arXiv2017

TernGrad: Ternary Gradients to Reduce Communication in Distributed Deep Learning

High network communication cost for synchronizing gradients and parameters is the well-known bottleneck of distributed training. In this work, we propose TernGrad that uses ternary gradients to accelerate distributed deep learning in data parallelism. Our approach requires only three numerical levels {-1,0,1}, which can aggressively reduce the communication time. We mathematically prove the convergence of TernGrad under the assumption of a bound on gradients. Guided by the bound, we propose layer-wise ternarizing and gradient clipping to improve its convergence. Our experiments show that applying TernGrad on AlexNet does not incur any accuracy loss and can even improve accuracy. The accuracy loss of GoogLeNet induced by TernGrad is less than 2% on average. Finally, a performance model is proposed to study the scalability of TernGrad. Experiments show significant speed gains for various deep neural networks. Our source code is available.

cs.LG↗

arXiv2017

Coordinating Filters for Faster Deep Neural Networks

Very large-scale Deep Neural Networks (DNNs) have achieved remarkable successes in a large variety of computer vision tasks. However, the high computation intensity of DNNs makes it challenging to deploy these models on resource-limited systems. Some studies used low-rank approaches that approximate the filters by low-rank basis to accelerate the testing. Those works directly decomposed the pre-trained DNNs by Low-Rank Approximations (LRA). How to train DNNs toward lower-rank space for more efficient DNNs, however, remains as an open area. To solve the issue, in this work, we propose Force Regularization, which uses attractive forces to enforce filters so as to coordinate more weight information into lower-rank space. We mathematically and empirically verify that after applying our technique, standard LRA methods can reconstruct filters using much lower basis and thus result in faster DNNs. The effectiveness of our approach is comprehensively evaluated in ResNets, AlexNet, and GoogLeNet. In AlexNet, for example, Force Regularization gains 2x speedup on modern GPU without accuracy loss and 4.05x speedup on CPU by paying small accuracy degradation. Moreover, Force Regularization better initializes the low-rank DNNs such that the fine-tuning can converge faster toward higher accuracy. The obtained lower-rank DNNs can be further sparsified, proving that Force Regularization can be integrated with state-of-the-art sparsity-based acceleration methods. Source code is available in https://github.com/wenwei202/caffe

cs.CV↗

arXiv2003

SWIRE: The SIRTF Wide-area InfraRed Extragalactic Survey

The SIRTF Wide-area InfraRed Extragalactic survey (SWIRE), the largest SIRTF Legacy program, is a wide-area, imaging survey to trace the evolution of dusty, star-forming galaxies, evolved stellar populations, and AGN as a function of environment, from redshifts z~3 to the current epoch. SWIRE will survey 7 high-latitude fields, totaling 60 - 65 sq. deg. in all 7 SIRTF bands: IRAC 3.6, 4.5, 5.6, 8 microns and MIPS 24, 70, 160 microns. The Legacy Extragalactic Catalog may contain in excess of 2 million IR-selected galaxies, dominated by (1) ~150,000 luminous infrared galaxies (LIRGs: L{FIR}>10^11 L_sun), ~7000 of these with z>2; (2) 1 million early-type galaxies, ~10,000 with z>2; and (3) \~20,000 classical AGN, plus significantly more dust-obscured QSO/AGN among the LIRGs. SWIRE will provide an unprecedented view of the evolution of galaxies, structure, and AGN. The key scientific goals of SWIRE are: (1) to determine the evolution of actively star-forming and passively evolving galaxies in order to understand the history of galaxy formation in the context of cosmic structure formation; (2) to determine the evolution of the spatial distribution and clustering of evolved galaxies, starbursts and AGN in the key redshift range, 0.5<z<3, over which much of cosmic evolution has occurred; (3) to determine the evolutionary relationship between ``normal galaxies'' and AGN, and the contribution of AGN accretion energy vs stellar nucleosynthesis to the cosmic backgrounds. SWIRE's large area is important to create statistically significant population samples over enough volume cells that we can resolve the star formation history as a function of epoch and environment The large volume is also optimised for finding rare objects.

astro-ph↗

arXiv2002

Far-Infrared photometry of a statistical sampleof late-type Virgo Cluster galaxies

We present deep diffraction-limited far-infrared (FIR) strip maps of a sample of 63 galaxies later than S0 and brighter than B_T 16.8, selected from the Virgo Cluster Catalogue of Binggeli, Sandage & Tammann. The ISOPHOT instrument on board the Infrared Space Observatory was used to achieve sensitivities typically an order of magnitude deeper than IRAS in the 60 and 100 micron bands and to reach the confusion limit at 170 microns. The averaged 3 sigma upper limits for integrated flux densities of point sources at 60, 100 and 170 microns are 43, 33 and 58 mJy, respectively. 54 galaxies (85.7 percent) are detected at least at one wavelength, and 40 galaxies (63.5 percent) are detected at all three wavelengths. The highest detection rate (85.7 percent) is in the 170 micron band. In many cases the galaxies are resolved, allowing the scale length of the infrared disks to be derived from the oversampled brightness profiles in addition to the spatially integrated emission. The data presented should provide the basis for a variety of statistical investigations of the FIR spectral energy distributions of gas rich galaxies in the local universe spanning a broad range in star-formation activity and morphological types, including dwarf systems and galaxies with rather quiescent star formation activity.

astro-ph↗

arXiv2001

Models for Multiband IR Surveys

Empirical 'backward' galaxy evolution models for IR-bright galaxies are constrained using multiband IR surveys. A new Monte-Carlo algorithm is developed for this task. It exploits a large library of realistic Spectral Energy Distributions (SEDs) of 837 local IR galaxies (IRAS 25$μm$ selected) from the UV (1000Å) to the radio (20cm), including ISO-measured 3--13$μm$ unidentified broad features (UIBs). The basic assumption is that the local correlation between SEDs and Mid-Infrared (MIR) luminosities can be applied to earlier epochs of the Universe. Three populations of IR sources are considered in the evolution models. These include (1) starburst galaxies, (2) normal late-type galaxies, and (3) galaxies with AGN. A set of models so constructed are compared with data from the literature. Predictions for number counts, confusion limits, redshift distributions, and color-color diagrams are made for multiband surveys using the upcoming SIRTF satellite.

astro-ph↗

arXiv2000

Molecular Gas Concentrations Outside the Merging Disks

We present our preliminary BIMA CO(1-0) images of II~Zw~96 which show huge molecular gas concentrations outside the merging disks. The dominant extra-disk CO concentrations in II~Zw~96 correspond to the two star-forming ``knots'' hidden by dust, whereas the others are associated with the two spiral disks. The unique huge CO concentrations, $\approxgt 5$ kpc away from the merging nuclei, account for $\approxgt 50 %$ of the total CO emission, yet only at a moderate gas surface density as compared with that of ultraluminous infrared galaxies (ULIGs). We tentatively conclude that the huge extra-disk gas concentrations are most likely the overlap region of the gas disks, not necessarily coinciding with the stellar disks overlap region. The gas disks overlap region could be formed prior to the merging of the stellar disks and become the overlap regions found between the disks like that in Arp~244 (the ``Antennae'' galaxies). These overlap regions would eventually nurture the {\it extreme starbursts} and evolve into the extended gas disk surrounding the double nuclei in final merging in most ULIGs.

astro-ph↗

arXiv2000

CO in Stephan's Quintet: First Evidence of Molecular Gas in the Intragroup Starburst

We present the first interferometric evidence of CO(1-0) emission in the intragroup starburst (IGS) region of HCG~92 (Stephan's Quintet). BIMA's large primary beam covers fully both the IGS region and the dominant group member galaxy Seyfert NGC 7319, as well as partially the intruder galaxy NGC 7318B. CO emission of $\approxgt 4σ$ is detected in all of them. The detection of the CO emission associated with the IGS is further supported by comparisons with observations in other wavebands (e.g., HI, mid-IR, \ha). Assuming the standard conversion factor, 3.6--8.0$\times 10^8~\Msun$ molecular gas is found to be associated with the IGS. Confining to the region of the CO emission peak, a relatively high star-formation efficiency of 8.6$\Lsun/\Msun$ is derived for the IGS, which is nearly comparable to that of local starburst galaxies (e.g., M82).

astro-ph↗