arXiv subjects

Xiang Li

Publications and source records attributed to Xiang Li.

At least 1,171 records · Page 65Linked to original sources

arXiv2021

Automated Pulmonary Embolism Detection from CTPA Images Using an End-to-End Convolutional Neural Network

Automated methods for detecting pulmonary embolisms (PEs) on CT pulmonary angiography (CTPA) images are of high demand. Existing methods typically employ separate steps for PE candidate detection and false positive removal, without considering the ability of the other step. As a result, most existing methods usually suffer from a high false positive rate in order to achieve an acceptable sensitivity. This study presents an end-to-end trainable convolutional neural network (CNN) where the two steps are optimized jointly. The proposed CNN consists of three concatenated subnets: 1) a novel 3D candidate proposal network for detecting cubes containing suspected PEs, 2) a 3D spatial transformation subnet for generating fixed-sized vessel-aligned image representation for candidates, and 3) a 2D classification network which takes the three cross-sections of the transformed cubes as input and eliminates false positives. We have evaluated our approach using the 20 CTPA test dataset from the PE challenge, achieving a sensitivity of 78.9%, 80.7% and 80.7% at 2 false positives per volume at 0mm, 2mm and 5mm localization error, which is superior to the state-of-the-art methods. We have further evaluated our system on our own dataset consisting of 129 CTPA data with a total of 269 emboli. Our system achieves a sensitivity of 63.2%, 78.9% and 86.8% at 2 false positives per volume at 0mm, 2mm and 5mm localization error.

cs.CV↗

arXiv2021

Spatial-spectral Terahertz Networks

This paper focuses on the spatial-spectral terahertz (THz) networks, where transmitters equipped with leaky-wave antennas send information to their receivers at the THz frequency bands. As a directional and nearly planar antenna, the leaky-wave antenna allows for information transmissions with narrow beams and high antenna gains. The conventional large antenna arrays are confronted with challenging issues such as scaling limits and path discovery in the THz frequencies. Therefore, this work exploits the potential of leaky-wave antennas in the dense THz networks, to establish low-complexity THz links. By addressing the propagation angle-frequency coupling effects, the transmission rate is analyzed. The results show that the leaky-wave antenna is efficient for achieving the high-speed transmission rate. The co-channel interference management is unnecessary when the THz transmitters with large subchannel bandwidths are not extremely dense. A simple subchannel allocation solution is proposed, which enhances the transmission rate compared with the same number of subchannels with the equal allocation of the frequency band. After subchannel allocation, a low-complexity power allocation method is proposed to improve the energy efficiency.

cs.IT↗

arXiv2021

Improved Loss Function-Based Prediction Method of Extreme Temperatures in Greenhouses

The prediction of extreme greenhouse temperatures to which crops are susceptible is essential in the field of greenhouse planting. It can help avoid heat or freezing damage and economic losses. Therefore, it's important to develop models that can predict them accurately. Due to the lack of extreme temperature data in datasets, it is challenging for models to accurately predict it. In this paper, we propose an improved loss function, which is suitable for a variety of machine learning models. By increasing the weight of extreme temperature samples and reducing the possibility of misjudging extreme temperature as normal, the proposed loss function can enhance the prediction results in extreme situations. To verify the effectiveness of the proposed method, we implement the improved loss function in LightGBM, long short-term memory, and artificial neural network and conduct experiments on a real-world greenhouse dataset. The results show that the performance of models with the improved loss function is enhanced compared to the original models in extreme cases. The improved models can be used to guarantee the timely judgment of extreme temperatures in agricultural greenhouses, thereby preventing unnecessary losses caused by incorrect predictions.

cs.AI↗

arXiv2021

How does homophily shape the topology of a dynamic network?

We consider a dynamic network of individuals that may hold one of two different opinions in a two-party society. As a dynamical model, agents can endlessly create and delete links to satisfy a preferred degree, and the network is shaped by \emph{homophily}, a form of social interaction. Characterized by the parameter $J \in [-1,1]$, the latter plays a role similar to Ising spins: agents create links to others of the same opinion with probability $(1+J)/2$, and delete them with probability $(1-J)/2$. Using Monte Carlo simulations and mean-field theory, we focus on the network structure in the steady state. We study the effects of $J$ on degree distributions and the fraction of cross-party links. While the extreme cases of homophily or heterophily ($J= \pm 1$) are easily understood to result in complete polarization or anti-polarization, intermediate values of $J$ lead to interesting features of the network. Our model exhibits the intriguing feature of an "overwhelming transition" occurring when communities of different sizes are subject to sufficient heterophily: agents of the minority group are oversubscribed and their average degree greatly exceeds that of the majority group. In addition, we introduce an original measure of polarization which displays distinct advantages over the commonly used average edge homogeneity.

cond-mat.stat-mech↗

arXiv2021

Magnetic order, disorder, and excitations under pressure in the Mott insulator Sr$_2$IrO$_4$

Protected by the interplay of on-site Coulomb interactions and spin-orbit coupling, Sr$_2$IrO$_4$ at high pressure is a rare example of a Mott insulator with a paramagnetic ground state. Here, using optical Raman scattering, we measure both the phonon and magnon evolution in Sr$_2$IrO$_4$ under pressure, and identify three different magnetically-ordered phases, culminating in a spin-disordered state beyond 18 GPa. A strong first-order structural phase transition drives the magnetic evolution at $\sim$10 GPa with reduced structural anisotropy in the IrO$_6$ cages, leading to increasingly isotropic exchange interactions between the Heisenberg spins and a spin-flip transition to $c$-axis-aligned antiferromagnetic order. In the disordered phase of Heisenberg $J_\mathrm{eff}=1/2$ pseudospins, the spin excitations are quasi-elastic and continuous to 10 meV, potentially hosting a gapless quantum spin liquid in Sr$_2$IrO$_4$.

cond-mat.str-el↗

arXiv2021

Unified Style Transfer

Currently, it is hard to compare and evaluate different style transfer algorithms due to chaotic definitions of style and the absence of agreed objective validation methods in the study of style transfer. In this paper, a novel approach, the Unified Style Transfer (UST) model, is proposed. With the introduction of a generative model for internal style representation, UST can transfer images in two approaches, i.e., Domain-based and Image-based, simultaneously. At the same time, a new philosophy based on the human sense of art and style distributions for evaluating the transfer model is presented and demonstrated, called Statistical Style Analysis. It provides a new path to validate style transfer models' feasibility by validating the general consistency between internal style representation and art facts. Besides, the translation-invariance of AdaIN features is also discussed.

cs.CV↗

arXiv2021

Video Instance Segmentation by Instance Flow Assembly

Instance segmentation is a challenging task aiming at classifying and segmenting all object instances of specific classes. While two-stage box-based methods achieve top performances in the image domain, they cannot easily extend their superiority into the video domain. This is because they usually deal with features or images cropped from the detected bounding boxes without alignment, failing to capture pixel-level temporal consistency. We embrace the observation that bottom-up methods dealing with box-free features could offer accurate spacial correlations across frames, which can be fully utilized for object and pixel level tracking. We first propose our bottom-up framework equipped with a temporal context fusion module to better encode inter-frame correlations. Intra-frame cues for semantic segmentation and object localization are simultaneously extracted and reconstructed by corresponding decoders after a shared backbone. For efficient and robust tracking among instances, we introduce an instance-level correspondence across adjacent frames, which is represented by a center-to-center flow, termed as instance flow, to assemble messy dense temporal correspondences. Experiments demonstrate that the proposed method outperforms the state-of-the-art online methods (taking image-level input) on the challenging Youtube-VIS dataset.

cs.CV↗

arXiv2021

The Image Local Autoregressive Transformer

Recently, AutoRegressive (AR) models for the whole image generation empowered by transformers have achieved comparable or even better performance to Generative Adversarial Networks (GANs). Unfortunately, directly applying such AR models to edit/change local image regions, may suffer from the problems of missing global information, slow inference speed, and information leakage of local guidance. To address these limitations, we propose a novel model -- image Local Autoregressive Transformer (iLAT), to better facilitate the locally guided image synthesis. Our iLAT learns the novel local discrete representations, by the newly proposed local autoregressive (LA) transformer of the attention mask and convolution mechanism. Thus iLAT can efficiently synthesize the local image regions by key guidance information. Our iLAT is evaluated on various locally guided image syntheses, such as pose-guided person image synthesis and face editing. Both the quantitative and qualitative results show the efficacy of our model.

cs.CV↗

arXiv2021

Student Helping Teacher: Teacher Evolution via Self-Knowledge Distillation

Knowledge distillation usually transfers the knowledge from a pre-trained cumbersome teacher network to a compact student network, which follows the classical teacher-teaching-student paradigm. Based on this paradigm, previous methods mostly focus on how to efficiently train a better student network for deployment. Different from the existing practices, in this paper, we propose a novel student-helping-teacher formula, Teacher Evolution via Self-Knowledge Distillation (TESKD), where the target teacher (for deployment) is learned with the help of multiple hierarchical students by sharing the structural backbone. The diverse feedback from multiple students allows the teacher to improve itself through the shared feature representations. The effectiveness of our proposed framework is demonstrated by extensive experiments with various network settings on two standard benchmarks including CIFAR-100 and ImageNet. Notably, when trained together with our proposed method, ResNet-18 achieves 79.15% and 71.14% accuracy on CIFAR-100 and ImageNet, outperforming the baseline results by 4.74% and 1.43%, respectively. The code is available at: https://github.com/zhengli427/TESKD.

cs.CV↗

arXiv2021

Feasible Architecture for Quantum Fully Convolutional Networks

Fully convolutional networks are robust in performing semantic segmentation, with many applications from signal processing to computer vision. From the fundamental principles of variational quantum algorithms, we propose a feasible pure quantum architecture that can be operated on noisy intermediate-scale quantum devices. In this work, a parameterized quantum circuit consisting of three layers, convolutional, pooling, and upsampling, is characterized by generative one-qubit and two-qubit gates and driven by a classical optimizer. This architecture supplies a solution for realizing the dynamical programming on a one-way quantum computer and maximally taking advantage of quantum computing throughout the calculation. Moreover, our algorithm works on many physical platforms, and particularly the upsampling layer can use either conventional qubits or multiple-level systems. Through numerical simulations, our study represents the successful training of a pure quantum fully convolutional network and discusses advantages by comparing it with the hybrid solution.

quant-ph↗

arXiv2021

Observations of Forbush Decreases of cosmic ray electrons and positrons with the Dark Matter Particle Explorer

The Forbush Decrease (FD) represents the rapid decrease of the intensities of charged particles accompanied with the coronal mass ejections (CMEs) or high-speed streams from coronal holes. It has been mainly explored with ground-based neutron monitors network which indirectly measure the integrated intensities of all species of cosmic rays by counting secondary neutrons produced from interaction between atmosphere atoms and cosmic rays. The space-based experiments can resolve the species of particles but the energy ranges are limited by the relative small acceptances except for the most abundant particles like protons and helium. Therefore, the FD of cosmic ray electrons and positrons have just been investigated by the PAMELA experiment in the low energy range ($<5$ GeV) with limited statistics. In this paper, we study the FD event occurred in September, 2017, with the electron and positron data recorded by the Dark Matter Particle Explorer. The evolution of the FDs from 2 GeV to 20 GeV with a time resolution of 6 hours are given. We observe two solar energetic particle events in the time profile of the intensity of cosmic rays, the earlier and weak one has not been shown in the neutron monitor data. Furthermore, both the amplitude and recovery time of fluxes of electrons and positrons show clear energy-dependence, which is important in probing the disturbances of the interplanetary environment by the coronal mass ejections.

astro-ph.HE↗

arXiv2021

Myopic Bike and Say Hi: Games for Empathizing with The Myopic

Myopia is an eye condition that makes it difficult for people to focus on faraway objects. It has become one of the most serious eye conditions worldwide and negatively impacts the quality of life of those who suffer from it. Although myopia is prevalent, many non-myopic people have misconceptions about it and encounter challenges empathizing with myopia situations and those who suffer from it. In this research, we developed two virtual reality (VR) games, (1) Myopic Bike and (2) Say Hi, to provide a means for the non-myopic population to experience the frustration and difficulties of myopic people. Our two games simulate two inconvenient daily life scenarios (riding a bicycle and greeting someone on the street) that myopic people encounter when not wearing glasses. We evaluated four participants' game experiences through questionnaires and semi-structured interviews. Overall, our two VR games can create an engaging and non-judgmental experience for the non-myopic population to better understand and empathize with those who suffer from myopia.

cs.HC↗

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

A kilonova from an ultra-quick merger of a neutron star binary

GRB 060505 was the first well-known nearby (at redshift 0.089) "hybrid" gamma-ray burst that has a duration longer than 2 seconds but without the association of a supernova down to very stringent limits. The prompt $γ-$ray flash lasting $\sim 4$ sec could consist of an intrinsic short burst and its tail emission, but the sizable temporal lag ($\sim 0.35$ sec) as well as the environment properties led to the widely-accepted classification of a long duration gamma-ray burst originated from the collapse of a massive star. Here for the $ first$ time we report the convincing evidence for a thermal-like optical radiation component in the spectral energy distribution of the early afterglow emission. In comparison to AT2017gfo, the thermal radiation is $\sim 2$ times brighter and the temperature is comparable at similar epochs. The optical decline is much quicker than that in X-rays, which is also at odds with the fireball afterglow model but quite natural for the presence of a blue kilonova. Our finding reveals a neutron star merger origin of the hybrid GRB 060505 and strongly supports the theoretical speculation that some binary neutron stars can merge ultra-quickly (within $\sim 1$ Myr) after their formation when the surrounding region is still highly star-forming and the metallicity remains low. Gravitational wave and electromagnetic jointed observations are expected to confirm such scenarios in the near future.

astro-ph.HE↗

arXiv2021

Fine-Grained Few Shot Learning with Foreground Object Transformation

Traditional fine-grained image classification generally requires abundant labeled samples to deal with the low inter-class variance but high intra-class variance problem. However, in many scenarios we may have limited samples for some novel sub-categories, leading to the fine-grained few shot learning (FG-FSL) setting. To address this challenging task, we propose a novel method named foreground object transformation (FOT), which is composed of a foreground object extractor and a posture transformation generator. The former aims to remove image background, which tends to increase the difficulty of fine-grained image classification as it amplifies the intra-class variance while reduces inter-class variance. The latter transforms the posture of the foreground object to generate additional samples for the novel sub-category. As a data augmentation method, FOT can be conveniently applied to any existing few shot learning algorithm and greatly improve its performance on FG-FSL tasks. In particular, in combination with FOT, simple fine-tuning baseline methods can be competitive with the state-of-the-art methods both in inductive setting and transductive setting. Moreover, FOT can further boost the performances of latest excellent methods and bring them up to the new state-of-the-art. In addition, we also show the effectiveness of FOT on general FSL tasks.

cs.CV↗

arXiv2021

Cost-Effective Federated Learning in Mobile Edge Networks

Federated learning (FL) is a distributed learning paradigm that enables a large number of mobile devices to collaboratively learn a model under the coordination of a central server without sharing their raw data. Despite its practical efficiency and effectiveness, the iterative on-device learning process (e.g., local computations and global communications with the server) incurs a considerable cost in terms of learning time and energy consumption, which depends crucially on the number of selected clients and the number of local iterations in each training round. In this paper, we analyze how to design adaptive FL in mobile edge networks that optimally chooses these essential control variables to minimize the total cost while ensuring convergence. We establish the analytical relationship between the total cost and the control variables with the convergence upper bound. To efficiently solve the cost minimization problem, we develop a low-cost sampling-based algorithm to learn the convergence related unknown parameters. We derive important solution properties that effectively identify the design principles for different optimization metrics. Practically, we evaluate our theoretical results both in a simulated environment and on a hardware prototype. Experimental evidence verifies our derived properties and demonstrates that our proposed solution achieves near-optimal performance for different optimization metrics for various datasets and heterogeneous system and statistical settings.

cs.LG↗

arXiv2021

Fermionic Neural Network with Effective Core Potential

Deep learning techniques have opened a new venue for electronic structure theory in recent years. In contrast to traditional methods, deep neural networks provide much more expressive and flexible wave function ansatz, resulting in better accuracy and time scaling behavior. In order to study larger systems while retaining sufficient accuracy, we integrate a powerful neural-network based model (FermiNet) with the effective core potential method, which helps to reduce the complexity of the problem by replacing inner core electrons with additional semi-local potential terms in Hamiltonian. In this work, we calculate the ground state energy of 3d transition metal atoms and their monoxide which are quite challenging for original FermiNet work, and the results are in good consistency with both experimental data and other state-of-the-art computational methods. Our development is an important step for a broader application of deep learning in the electronic structure calculation of molecules and materials.

physics.chem-ph↗

arXiv2021

Electronic nature of chiral charge order in the kagome superconductor CsV3Sb5

Kagome superconductors with Tc up to 7K have been discovered over 40 years. Recently, unconventional chiral charge order has been reported in kagome superconductor KV3Sb5, with an ordering temperature of one order of magnitude higher than the TC. However, the chirality of the charge order has not been reported in the cousin kagome superconductor CsV3Sb5, and the electronic nature of the chirality remains elusive. In this letter, we report the observation of electronic chiral charge order in CsV3Sb5 via scanning tunneling microscopy (STM). We observe a 2x2 charge modulation and a 1x4 superlattice in both topographic data and tunneling spectroscopy. 2x2 charge modulation is highly anticipated as a charge order by fundamental kagome lattice models at van Hove filling, and is shown to exhibit intrinsic chirality. We find that the 1x4 superlattices forms various small domain walls, and can be a surface effect as supported by our first-principles calculations. Crucially, we find that the amplitude of the energy gap opened by the charge order exhibits real space modulations, and features 2x2 wave vectors with chirality, highlighting the electronic nature of the chiral charge order. STM study at 0.4K reveals a superconducting energy gap with a gap size 2Δ=0.85meV, which estimates a moderate superconductivity coupling strength with 2Δ/kBTc=3.9. When further applying a c-axis magnetic field, vortex core bound states are observed within this gap, indicative of clean-limit superconductivity.

cond-mat.str-el↗