arXiv subjects

Yu Zhang

Publications and source records attributed to Yu Zhang.

At least 1,387 records · Page 77Linked to original sources

arXiv2020

Residual Attention U-Net for Automated Multi-Class Segmentation of COVID-19 Chest CT Images

The novel coronavirus disease 2019 (COVID-19) has been spreading rapidly around the world and caused significant impact on the public health and economy. However, there is still lack of studies on effectively quantifying the lung infection caused by COVID-19. As a basic but challenging task of the diagnostic framework, segmentation plays a crucial role in accurate quantification of COVID-19 infection measured by computed tomography (CT) images. To this end, we proposed a novel deep learning algorithm for automated segmentation of multiple COVID-19 infection regions. Specifically, we use the Aggregated Residual Transformations to learn a robust and expressive feature representation and apply the soft attention mechanism to improve the capability of the model to distinguish a variety of symptoms of the COVID-19. With a public CT image dataset, we validate the efficacy of the proposed algorithm in comparison with other competing methods. Experimental results demonstrate the outstanding performance of our algorithm for automated segmentation of COVID-19 Chest CT images. Our study provides a promising deep leaning-based segmentation tool to lay a foundation to quantitative diagnosis of COVID-19 lung infection in CT images.

eess.IV↗

arXiv2020

Customized data-driven RANS closures for bi-fidelity LES-RANS optimization

Multi-fidelity optimization methods promise a high-fidelity optimum at a cost only slightly greater than a low-fidelity optimization. This promise is seldom achieved in practice, due to the requirement that low- and high-fidelity models correlate well. In this article, we propose an efficient bi-fidelity shape optimization method for turbulent fluid-flow applications with Large-Eddy Simulation (LES) and Reynolds-averaged Navier-Stokes (RANS) as the high- and low-fidelity models within a hierarchical-Kriging surrogate modelling framework. Since the LES-RANS correlation is often poor, we use the full LES flow-field at a single point in the design space to derive a custom-tailored RANS closure model that reproduces the LES at that point. This is achieved with machine-learning techniques, specifically sparse regression to obtain high corrections of the turbulence anisotropy tensor and the production of turbulence kinetic energy as functions of the RANS mean-flow. The LES-RANS correlation is dramatically improved throughout the design-space. We demonstrate the effectiveness and efficiency of our method in a proof-of-concept shape optimization of the well-known periodic-hill case. Standard RANS models perform poorly in this case, whereas our method converges to the LES-optimum with only two LES samples.

physics.comp-ph↗

arXiv2020

Transferring entangled states of photonic cat-state qubits in circuit QED

We propose a method for transferring quantum entangled states of two photonic cat-state qubits (cqubits) from two microwave cavities to the other two microwave cavities. This proposal is realized by using four microwave cavities coupled to a superconducting flux qutrit. Because of using four cavities with different frequencies, the inter-cavity crosstalk is significantly reduced. Since only one coupler qutrit is used, the circuit resources is minimized. The entanglement transfer is completed with a single-step operation only, thus this proposal is quite simple. The third energy level of the coupler qutrit is not populated during the state transfer, therefore decoherence from the higher energy level is greatly suppressed. Our numerical simulations show that high-fidelity transfer of two-cqubit entangled states from two transmission line resonators to the other two transmission line resonators is feasible with current circuit QED technology. This proposal is universal and can be applied to accomplish the same task in a wide range of physical systems, such as four microwave or optical cavities, which are coupled to a natural or artificial three-level atom.

quant-ph↗

arXiv2020

Generation of quantum entangled states of multiple groups of qubits distributed in multiple cavities

Provided that cavities are initially in a Greenberger-Horne-Zeilinger (GHZ) entangled state, we show that GHZ states of N-group qubits distributed in N cavities can be created via a 3-step operation. The GHZ states of the N-group qubits are generated by using N-group qutrits placed in the N cavities. Here, "qutrit" refers to a three-level quantum system with the two lowest levels representing a qubit while the third level acting as an intermediate state necessary for the GHZ state creation. This proposal does not depend on the architecture of the cavity-based quantum network and the way for coupling the cavities. The operation time is independent of the number of qubits. The GHZ states are prepared deterministically because no measurement on the states of qutrits or cavities is needed. In addition, the third energy level of the qutrits during the entire operation is virtually excited and thus decoherence from higher energy levels is greatly suppressed. This proposal is quite general and can in principle be applied to create GHZ states of many qubits using different types of physical qutrits (e.g., atoms, quantum dots, NV centers, various superconducting qutrits, etc.) distributed in multiple cavities. As a specific example, we further discuss the experimental feasibility of preparing a GHZ state of four-group transmon qubits (each group consisting of three qubits) distributed in four one-dimensional transmission line resonators arranged in an array.

quant-ph↗

arXiv2020

Diversifying Seeds and Audience in Social Influence Maximization

Influence maximization (IM) has been extensively studied for better viral marketing. However, previous works put less emphasis on how balancedly the audience are affected across different communities and how diversely the seed nodes are selected. In this paper, we incorporate audience diversity and seed diversity into the IM task. From the model perspective, in order to characterize both influence spread and diversity in our objective function, we adopt three commonly used utilities in economics (i.e., Perfect Substitutes, Perfect Complements and Cobb-Douglas). We validate our choices of these three functions by showing their nice properties. From the algorithmic perspective, we present various approximation strategies to maximize the utilities. In audience diversification, we propose a solution-dependent approximation algorithm to circumvent the hardness results. In seed diversification, we prove a ($1/e-ε$) approximation ratio based on non-monotonic submodular maximization. Experimental results show that our framework outperforms other natural heuristics both in utility maximization and result diversification.

cs.SI↗

arXiv2020

CF2-Net: Coarse-to-Fine Fusion Convolutional Network for Breast Ultrasound Image Segmentation

Breast ultrasound (BUS) image segmentation plays a crucial role in a computer-aided diagnosis system, which is regarded as a useful tool to help increase the accuracy of breast cancer diagnosis. Recently, many deep learning methods have been developed for segmentation of BUS image and show some advantages compared with conventional region-, model-, and traditional learning-based methods. However, previous deep learning methods typically use skip-connection to concatenate the encoder and decoder, which might not make full fusion of coarse-to-fine features from encoder and decoder. Since the structure and edge of lesion in BUS image are common blurred, these would make it difficult to learn the discriminant information of structure and edge, and reduce the performance. To this end, we propose and evaluate a coarse-to-fine fusion convolutional network (CF2-Net) based on a novel feature integration strategy (forming an 'E'-like type) for BUS image segmentation. To enhance contour and provide structural information, we concatenate a super-pixel image and the original image as the input of CF2-Net. Meanwhile, to highlight the differences in the lesion regions with variable sizes and relieve the imbalance issue, we further design a weighted-balanced loss function to train the CF2-Net effectively. The proposed CF2-Net was evaluated on an open dataset by using four-fold cross validation. The results of the experiment demonstrate that the CF2-Net obtains state-of-the-art performance when compared with other deep learning-based methods

eess.IV↗

arXiv2020

Fisher Deep Domain Adaptation

Deep domain adaptation models learn a neural network in an unlabeled target domain by leveraging the knowledge from a labeled source domain. This can be achieved by learning a domain-invariant feature space. Though the learned representations are separable in the source domain, they usually have a large variance and samples with different class labels tend to overlap in the target domain, which yields suboptimal adaptation performance. To fill the gap, a Fisher loss is proposed to learn discriminative representations which are within-class compact and between-class separable. Experimental results on two benchmark datasets show that the Fisher loss is a general and effective loss for deep domain adaptation. Noticeable improvements are brought when it is used together with widely adopted transfer criteria, including MMD, CORAL and domain adversarial loss. For example, an absolute improvement of 6.67% in terms of the mean accuracy is attained when the Fisher loss is used together with the domain adversarial loss on the Office-Home dataset.

cs.LG↗

arXiv2020

Speech Sentiment Analysis via Pre-trained Features from End-to-end ASR Models

In this paper, we propose to use pre-trained features from end-to-end ASR models to solve speech sentiment analysis as a down-stream task. We show that end-to-end ASR features, which integrate both acoustic and text information from speech, achieve promising results. We use RNN with self-attention as the sentiment classifier, which also provides an easy visualization through attention weights to help interpret model predictions. We use well benchmarked IEMOCAP dataset and a new large-scale speech sentiment dataset SWBD-sentiment for evaluation. Our approach improves the-state-of-the-art accuracy on IEMOCAP from 66.6% to 71.7%, and achieves an accuracy of 70.10% on SWBD-sentiment with more than 49,500 utterances.

cs.CL↗

arXiv2020

Sum Rate Optimization for Two Way Communications with Intelligent Reflecting Surface

In this letter, an intelligent reflecting surface (IRS) enhanced full-duplex MIMO two-way communication system is studied. The system sum rate is maximized through jointly optimizing the source precoders and the IRS phase shift matrix. Adopting the idea of Arimoto-Blahut algorithm, the non-convex optimization problem is decoupled into three sub-problems, which are solved alternatingly. All the sub-problems can be solved efficiently with closed-form solutions. In addition, practical IRS assumptions, e.g., discrete phase shift levels, are also considered. Numerical results verify the convergence and performance of the proposed scheme.

eess.SP↗

arXiv2020

Unsupervised Domain Adaptation for Mammogram Image Classification: A Promising Tool for Model Generalization

Generalization is one of the key challenges in the clinical validation and application of deep learning models to medical images. Studies have shown that such models trained on publicly available datasets often do not work well on real-world clinical data due to the differences in patient population and image device configurations. Also, manually annotating clinical images is expensive. In this work, we propose an unsupervised domain adaptation (UDA) method using Cycle-GAN to improve the generalization ability of the model without using any additional manual annotations.

cs.CV↗

arXiv2020

A singularity at the criticality for the free energy in percolation

Consider percolation on the triangular lattice. Let $κ(p)$ be the free energy at the zero field. We show that $$|κ'''(p)| \leq |p-p_c|^{-1/3+o(1)} \mbox{ if } p \neq p_c.$$ Furthermore, we show that there exists a sequence $ε_n\downarrow 0$ such that $$|κ'''(p_c\pm ε_n)|\geq ε_n^{-1/3+o(1)}.$$ This answers affirmatively a conjecture, asked by Sykes and Essam a half century ago, whether $κ(p)$ has a singularity at the criticality.

math.PR↗

arXiv2020

Defense-PointNet: Protecting PointNet Against Adversarial Attacks

Despite remarkable performance across a broad range of tasks, neural networks have been shown to be vulnerable to adversarial attacks. Many works focus on adversarial attacks and defenses on 2D images, but few focus on 3D point clouds. In this paper, our goal is to enhance the adversarial robustness of PointNet, which is one of the most widely used models for 3D point clouds. We apply the fast gradient sign attack method (FGSM) on 3D point clouds and find that FGSM can be used to generate not only adversarial images but also adversarial point clouds. To minimize the vulnerability of PointNet to adversarial attacks, we propose Defense-PointNet. We compare our model with two baseline approaches and show that Defense-PointNet significantly improves the robustness of the network against adversarial samples.

cs.CV↗

arXiv2020

2D Convolutional Neural Networks for 3D Digital Breast Tomosynthesis Classification

Automated methods for breast cancer detection have focused on 2D mammography and have largely ignored 3D digital breast tomosynthesis (DBT), which is frequently used in clinical practice. The two key challenges in developing automated methods for DBT classification are handling the variable number of slices and retaining slice-to-slice changes. We propose a novel deep 2D convolutional neural network (CNN) architecture for DBT classification that simultaneously overcomes both challenges. Our approach operates on the full volume, regardless of the number of slices, and allows the use of pre-trained 2D CNNs for feature extraction, which is important given the limited amount of annotated training data. In an extensive evaluation on a real-world clinical dataset, our approach achieves 0.854 auROC, which is 28.80% higher than approaches based on 3D CNNs. We also find that these improvements are stable across a range of model configurations.

cs.CV↗

arXiv2020

Joint 2D-3D Breast Cancer Classification

Breast cancer is the malignant tumor that causes the highest number of cancer deaths in females. Digital mammograms (DM or 2D mammogram) and digital breast tomosynthesis (DBT or 3D mammogram) are the two types of mammography imagery that are used in clinical practice for breast cancer detection and diagnosis. Radiologists usually read both imaging modalities in combination; however, existing computer-aided diagnosis tools are designed using only one imaging modality. Inspired by clinical practice, we propose an innovative convolutional neural network (CNN) architecture for breast cancer classification, which uses both 2D and 3D mammograms, simultaneously. Our experiment shows that the proposed method significantly improves the performance of breast cancer classification. By assembling three CNN classifiers, the proposed model achieves 0.97 AUC, which is 34.72% higher than the methods using only one imaging modality.

cs.CV↗

arXiv2020

Self-supervised Image Enhancement Network: Training with Low Light Images Only

This paper proposes a self-supervised low light image enhancement method based on deep learning. Inspired by information entropy theory and Retinex model, we proposed a maximum entropy based Retinex model. With this model, a very simple network can separate the illumination and reflectance, and the network can be trained with low light images only. We introduce a constraint that the maximum channel of the reflectance conforms to the maximum channel of the low light image and its entropy should be largest in our model to achieve self-supervised learning. Our model is very simple and does not rely on any well-designed data set (even one low light image can complete the training). The network only needs minute-level training to achieve image enhancement. It can be proved through experiments that the proposed method has reached the state-of-the-art in terms of processing speed and effect.

cs.CV↗

arXiv2020

Learning Beam Codebooks with Neural Networks: Towards Environment-Aware mmWave MIMO

Scaling the number of antennas up is a key characteristic of current and future wireless communication systems. The hardware cost and power consumption, however, motivate large-scale MIMO systems, especially at millimeter wave (mmWave) bands, to rely on analog-only or hybrid analog/digital transceiver architectures. With these architectures, mmWave base stations normally use pre-defined beamforming codebooks for both initial access and data transmissions. Current beam codebooks, however, generally adopt single-lobe narrow beams and scan the entire angular space. This leads to high beam training overhead and loss in the achievable beamforming gains. In this paper, we propose a new machine learning framework for learning beamforming codebooks in hardware-constrained large-scale MIMO systems. More specifically, we develop a neural network architecture that accounts for the hardware constraints and learns beam codebooks that adapt to the surrounding environment and the user locations. Simulation results highlight the capability of the proposed solution in learning multi-lobe beams and reducing the codebook size, which leads to noticeable gains compared to classical codebook design approaches.

cs.IT↗

arXiv2020

Deep Reinforcement Learning for Intelligent Reflecting Surfaces: Towards Standalone Operation

The promising coverage and spectral efficiency gains of intelligent reflecting surfaces (IRSs) are attracting increasing interest. In order to realize these surfaces in practice, however, several challenges need to be addressed. One of these main challenges is how to configure the reflecting coefficients on these passive surfaces without requiring massive channel estimation or beam training overhead. Earlier work suggested leveraging supervised learning tools to design the IRS reflection matrices. While this approach has the potential of reducing the beam training overhead, it requires collecting large datasets for training the neural network models. In this paper, we propose a novel deep reinforcement learning framework for predicting the IRS reflection matrices with minimal training overhead. Simulation results show that the proposed online learning framework can converge to the optimal rate that assumes perfect channel knowledge. This represents an important step towards realizing a standalone IRS operation, where the surface configures itself without any control from the infrastructure.

cs.IT↗

arXiv2020

Electron interactions in strain-induced zero-energy flat band in twisted bilayer graphene near the magic angle

In the vicinity of the magic angle in twisted bilayer graphene (TBG), the two low-energy van Hove singularities (VHSs) become exceedingly narrow1-10 and many exotic correlated states, such as superconductivity, ferromagnetism, and topological phases, are observed11-16. Heterostrain, which is almost unavoidable in the TBG, can modify its single-particle band structure and lead to novel properties of the TBG that have never been considered so far. Here, we show that heterostrain in a TBG near the magic angle generates a new zero-energy flat band between the two VHSs. Doping the TBG to partially fill the zero-energy flat band, we observe a correlation-induced gap of about 10 meV that splits the flat band. By applying perpendicular magnetic fields, a large and linear response of the gap to magnetic fields is observed, attributing to the emergence of large orbital magnetic moments in the TBG when valley degeneracy of the flat band is lifted by electron-electron interactions. The orbital magnetic moment per moire supercell is measured as about 15 uB in the TBG.

cond-mat.str-el↗