arXiv subjects

Yu Liu

Publications and source records attributed to Yu Liu.

At least 919 records · Page 51Linked to original sources

arXiv2020

Anisotropic magnetocaloric effect and critical behavior in CrSbSe$_3$

We report anisotropic magnetocaloric effect and critical behavior in quasi-one-dimensional ferromagnetic CrSbSe$_3$ single crystal. The maximum magnetic entropy change $-ΔS_M^{max}$ is 2.16 J kg$^{-1}$ K$^{-1}$ for easy $a$ axis (2.03 J kg$^{-1}$ K$^{-1}$ for hard $b$ axis) and the relative cooling power $RCP$ is 163.1 J kg$^{-1}$ for easy $a$ axis (142.1 J kg$^{-1}$ for hard $b$ axis) near $T_c$ with a magnetic field change of 50 kOe. The magnetocrystalline anisotropy constant $K_u$ is estimated to be 148.5 kJ m$^{-3}$ at 10 K, decreasing to 39.4 kJ m$^{-3}$ at 70 K. The rescaled $ΔS_M(T,H)$ curves along all three axes collapse onto a universal curve, respectively, confirming the second order ferromagnetic transition. Further critical behavior analysis around $T_c \sim$ 70 K gives that the critical exponents $β$ = 0.26(1), $γ$ = 1.32(2), and $δ$ = 6.17(9) for $H\parallel a$, while $β$ = 0.28(2), $γ$ = 1.02(1), and $δ$ = 4.14(16) for $H\parallel b$. The determined critical exponents suggest that the anisotropic magnetic coupling in CrSbSe$_3$ is strongly dependent on orientations of the applied magnetic field.

cond-mat.str-el↗

arXiv2020

Anisotropic magnetocaloric effect and critical behavior in CrCl$_3$

We report anisotropic magnetocaloric effect and magnetic critical behavior in van der Waals crystal CrCl$_3$. The maximum magnetic entropy change $-ΔS_M^{max} \sim 14.6$ J kg$^{-1}$ K$^{-1}$ and the relative cooling power $RCP \sim 340.3$ J kg$^{-1}$ near $T_c$ with a magnetic field change of 5 T are much larger when compared to CrI$_{3}$ or CrBr$_{3}$. The rescaled $ΔS_M(T,H)$ curves collapse onto a universal curve, confirming the second order ferromagnetic transition. Further critical behavior analysis around $T_c$ presents a set of critical exponents $β$ = 0.28(1) with $T_c$ = 19.4(2) K, $γ$ = 0.89(1) with $T_c$ = 18.95(8) K, and $δ$ = 4.6(1) at $T_c$ = 19 K, which are close to those of theoretical tricritical mean field model.

cond-mat.mtrl-sci↗

arXiv2020

Relative rigid subcategories and $τ$-tilting theory

Let $\mathcal B$ be an extriangulated category with enough projectives $\mathcal P$ and enough injectives $\mathcal I$, and let $\mathcal R$ be a contravariantly finite rigid subcategory of $\mathcal B$ which contains $\mathcal P$. We have an abelian quotient category $\mathcal H/\mathcal R\subseteq \mathcal B/\mathcal R$ which is equivalent ${\rm mod}(\mathcal R/\mathcal P)$. In this article, we find a one-to-one correspondence between support $τ$-tilting (resp. $τ$-rigid) subcategories of $\mathcal H/\mathcal R$ and maximal relative rigid (resp. relative rigid) subcategories of $\mathcal H$, and show that support tilting subcategories in $\mathcal H/\mathcal R$ is a special kind of support $τ$-tilting subcategories. We also study the relation between tilting subcategories of $\mathcal B/\mathcal R$ and cluster tilting subcategories of $\mathcal B$ when $\mathcal R$ is cluster tilting.

math.RT↗

arXiv2020

Generalizing Tensor Decomposition for N-ary Relational Knowledge Bases

With the rapid development of knowledge bases (KBs), link prediction task, which completes KBs with missing facts, has been broadly studied in especially binary relational KBs (a.k.a knowledge graph) with powerful tensor decomposition related methods. However, the ubiquitous n-ary relational KBs with higher-arity relational facts are paid less attention, in which existing translation based and neural network based approaches have weak expressiveness and high complexity in modeling various relations. Tensor decomposition has not been considered for n-ary relational KBs, while directly extending tensor decomposition related methods of binary relational KBs to the n-ary case does not yield satisfactory results due to exponential model complexity and their strong assumptions on binary relations. To generalize tensor decomposition for n-ary relational KBs, in this work, we propose GETD, a generalized model based on Tucker decomposition and Tensor Ring decomposition. The existing negative sampling technique is also generalized to the n-ary case for GETD. In addition, we theoretically prove that GETD is fully expressive to completely represent any KBs. Extensive evaluations on two representative n-ary relational KB datasets demonstrate the superior performance of GETD, significantly improving the state-of-the-art methods by over 15\%. Moreover, GETD further obtains the state-of-the-art results on the benchmark binary relational KB datasets.

cs.CL↗

arXiv2020

1st place solution for AVA-Kinetics Crossover in AcitivityNet Challenge 2020

This technical report introduces our winning solution to the spatio-temporal action localization track, AVA-Kinetics Crossover, in ActivityNet Challenge 2020. Our entry is mainly based on Actor-Context-Actor Relation Network. We describe technical details for the new AVA-Kinetics dataset, together with some experimental results. Without any bells and whistles, we achieved 39.62 mAP on the test set of AVA-Kinetics, which outperforms other entries by a large margin. Code will be available at: https://github.com/Siyu-C/ACAR-Net.

cs.CV↗

arXiv2020

Kondo scenario of the γ-α phase transition in single crystalline Cerium thin films

The physical mechanism driving the $γ$-$α$ phase transition of face-centre-cubic (fcc) cerium (Ce) remains controversial until now. In this work, high quality single crystalline fcc-Ce thin films were grown on Graphene/6$H$-SiC(0001) substrate, and explored by XRD and ARPES measurement. XRD spectra showed a clear $γ$-$α$ phase transition at $T_{γ-α}\approx$ 50 K, which is retarded by strain effect from substrate comparing with $T_{γ-α}$ (about 140 K) of the bulk Ce metal. However, APRES spectra did not show any signature of $α$-phase emerging in the surface-layer from 300 K to 17 K, which implied that $α$-phase might form at the bulk-layer of our Ce thin films. Besides, an evident Kondo dip near Fermi energy was observed in the APRES spectrum at 80 K, indicting the formation of Kondo singlet states in $γ$-Ce. Furthermore, the DFT+DMFT calculations were performed to simulate the electronic structures and the theoretical spectral functions agreed well with the experimental ARPES spectra. In $γ$-Ce, the behavior of the self-energy's imaginary part at low frequency not only confirmed that the Kondo singlet states emerged at $T_{\rm KS} \geq 80$ K, but also implied that they became coherent states at a lower characteristic temperature ($T_{\rm coh}\sim 40$ K) due to the indirect RKKY interaction among $f$-$f$ electrons. Besides, $T_{\rm coh}$ from the theoretical simulation was close to $T_{γ-α}$ from the XRD spectra. These issues suggested that the Kondo scenario might play an important role in the $γ$-$α$ phase transition of cerium thin films.

cond-mat.str-el↗

arXiv2020

PulseGAN: Learning to generate realistic pulse waveforms in remote photoplethysmography

Remote photoplethysmography (rPPG) is a non-contact technique for measuring cardiac signals from facial videos. High-quality rPPG pulse signals are urgently demanded in many fields, such as health monitoring and emotion recognition. However, most of the existing rPPG methods can only be used to get average heart rate (HR) values due to the limitation of inaccurate pulse signals. In this paper, a new framework based on generative adversarial network, called PulseGAN, is introduced to generate realistic rPPG pulse signals through denoising the chrominance signals. Considering that the cardiac signal is quasi-periodic and has apparent time-frequency characteristics, the error losses defined in time and spectrum domains are both employed with the adversarial loss to enforce the model generating accurate pulse waveforms as its reference. The proposed framework is tested on the public UBFC-RPPG database in both within-database and cross-database configurations. The results show that the PulseGAN framework can effectively improve the waveform quality, thereby enhancing the accuracy of HR, the heart rate variability (HRV) and the interbeat interval (IBI). The proposed method achieves the best performance compared to the denoising autoencoder (DAE) and CHROM, with the mean absolute error of AVNN (the average of all normal-to-normal intervals) improving 20.85% and 41.19%, and the mean absolute error of SDNN (the standard deviation of all NN intervals) improving 20.28% and 37.53%, respectively, in the cross-database test. This framework can be easily extended to other existing deep learning based rPPG methods, which is expected to expand the application scope of rPPG techniques.

eess.IV↗

arXiv2020

From $n$-exangulated categories to $n$-abelian categories

Herschend-Liu-Nakaoka introduced the notion of $n$-exangulated categories. It is not only a higher dimensional analogue of extriangulated categories defined by Nakaoka-Palu, but also gives a simultaneous generalization of $n$-exact categories in the sense of Jasso and $(n+2)$-angulated in the sense of Geiss-Keller-Oppermann. Let $\mathscr C$ be an $n$-exangulated category with enough projectives and enough injectives, and $\mathscr X$ a cluster tilting subcategory of $\mathscr C$. In this article, we show that the quotient category $\mathscr C/\mathscr X$ is an $n$-abelian category. This extends a result of Zhou-Zhu for $(n+2)$-angulated categories. Moreover, it highlights new phenomena when it is applied to $n$-exact categories.

math.RT↗

arXiv2020

Frobenius $n$-exangulated categories

Herschend-Liu-Nakaoka introduced the notion of $n$-exangulated categories as higher dimensional analogues of extriangulated categories defined by Nakaoka-Palu. The class of $n$-exangulated categories contains $n$-exact categories and $(n+2)$-angulated categories as examples. In this article, we introduce a notion of Frobenius $n$-exangulated categories which are a generalization of Frobenius $n$-exact categories. We show that the stable category of a Frobenius $n$-exangulated category is an $(n+2)$-angulated category. As an application, this result generalizes the work by Jasso. We provide a class of $n$-exangulated categories which are neither $n$-exact categories nor $(n+2)$-angulated categories. Finally, we discuss an application of the main results and give some examples illustrating it.

math.RT↗

arXiv2020

Modeling Event Propagation via Graph Biased Temporal Point Process

Temporal point process is widely used for sequential data modeling. In this paper, we focus on the problem of modeling sequential event propagation in graph, such as retweeting by social network users, news transmitting between websites, etc. Given a collection of event propagation sequences, conventional point process model consider only the event history, i.e. embed event history into a vector, not the latent graph structure. We propose a Graph Biased Temporal Point Process (GBTPP) leveraging the structural information from graph representation learning, where the direct influence between nodes and indirect influence from event history is modeled respectively. Moreover, the learned node embedding vector is also integrated into the embedded event history as side information. Experiments on a synthetic dataset and two real-world datasets show the efficacy of our model compared to conventional methods and state-of-the-art.

cs.SI↗

arXiv2020

A First Look at Commercial 5G Performance on Smartphones

We conduct to our knowledge a first measurement study of commercial 5G performance on smartphones by closely examining 5G networks of three carriers (two mmWave carriers, one mid-band carrier) in three U.S. cities. We conduct extensive field tests on 5G performance in diverse urban environments. We systematically analyze the handoff mechanisms in 5G and their impact on network performance. We explore the feasibility of using location and possibly other environmental information to predict the network performance. We also study the app performance (web browsing and HTTP download) over 5G. Our study consumes more than 15 TB of cellular data. Conducted when 5G just made its debut, it provides a "baseline" for studying how 5G performance evolves, and identifies key research directions on improving 5G users' experience in a cross-layer manner. We have released the data collected from our study (referred to as 5Gophers) at https://fivegophers.umn.edu/www20.

cs.NI↗

arXiv2020

A novel multimodal approach for hybrid brain-computer interface

Brain-computer interface (BCI) technologies have been widely used in many areas. In particular, non-invasive technologies such as electroencephalography (EEG) or near-infrared spectroscopy (NIRS) have been used to detect motor imagery, disease, or mental state. It has been already shown in literature that the hybrid of EEG and NIRS has better results than their respective individual signals. The fusion algorithm for EEG and NIRS sources is the key to implement them in real-life applications. In this research, we propose three fusion methods for the hybrid of the EEG and NIRS-based brain-computer interface system: linear fusion, tensor fusion, and $p$th-order polynomial fusion. Firstly, our results prove that the hybrid BCI system is more accurate, as expected. Secondly, the $p$th-order polynomial fusion has the best classification results out of the three methods, and also shows improvements compared with previous studies. For a motion imagery task and a mental arithmetic task, the best detection accuracy in previous papers were 74.20\% and 88.1\%, whereas our accuracy achieved was 77.53\% and 90.19\% . Furthermore, unlike complex artificial neural network methods, our proposed methods are not as computationally demanding.

cs.HC↗

arXiv2020

Cost-effectiveness Analysis of Antiepidemic Policies and Global Situation Assessment of COVID-19

With a two-layer contact-dispersion model and data in China, we analyze the cost-effectiveness of three types of antiepidemic measures for COVID-19: regular epidemiological control, local social interaction control, and inter-city travel restriction. We find that: 1) intercity travel restriction has minimal or even negative effect compared to the other two at the national level; 2) the time of reaching turning point is independent of the current number of cases, and only related to the enforcement stringency of epidemiological control and social interaction control measures; 3) strong enforcement at the early stage is the only opportunity to maximize both antiepidemic effectiveness and cost-effectiveness; 4) mediocre stringency of social interaction measures is the worst choice. Subsequently, we cluster countries/regions into four groups based on their control measures and provide situation assessment and policy suggestions for each group.

physics.soc-ph↗

arXiv2020

Change Detection in Heterogeneous Optical and SAR Remote Sensing Images via Deep Homogeneous Feature Fusion

Change detection in heterogeneous remote sensing images is crucial for disaster damage assessment. Recent methods use homogenous transformation, which transforms the heterogeneous optical and SAR remote sensing images into the same feature space, to achieve change detection. Such transformations mainly operate on the low-level feature space and may corrupt the semantic content, deteriorating the performance of change detection. To solve this problem, this paper presents a new homogeneous transformation model termed deep homogeneous feature fusion (DHFF) based on image style transfer (IST). Unlike the existing methods, the DHFF method segregates the semantic content and the style features in the heterogeneous images to perform homogeneous transformation. The separation of the semantic content and the style in homogeneous transformation prevents the corruption of image semantic content, especially in the regions of change. In this way, the detection performance is improved with accurate homogeneous transformation. Furthermore, we present a new iterative IST (IIST) strategy, where the cost function in each IST iteration measures and thus maximizes the feature homogeneity in additional new feature subspaces for change detection. After that, change detection is accomplished accurately on the original and the transformed images that are in the same feature space. Real remote sensing images acquired by SAR and optical satellites are utilized to evaluate the performance of the proposed method. The experiments demonstrate that the proposed DHFF method achieves significant improvement for change detection in heterogeneous optical and SAR remote sensing images, in terms of both accuracy rate and Kappa index.

cs.CV↗

arXiv2020

Anisotropic Convolutional Networks for 3D Semantic Scene Completion

As a voxel-wise labeling task, semantic scene completion (SSC) tries to simultaneously infer the occupancy and semantic labels for a scene from a single depth and/or RGB image. The key challenge for SSC is how to effectively take advantage of the 3D context to model various objects or stuffs with severe variations in shapes, layouts and visibility. To handle such variations, we propose a novel module called anisotropic convolution, which properties with flexibility and power impossible for the competing methods such as standard 3D convolution and some of its variations. In contrast to the standard 3D convolution that is limited to a fixed 3D receptive field, our module is capable of modeling the dimensional anisotropy voxel-wisely. The basic idea is to enable anisotropic 3D receptive field by decomposing a 3D convolution into three consecutive 1D convolutions, and the kernel size for each such 1D convolution is adaptively determined on the fly. By stacking multiple such anisotropic convolution modules, the voxel-wise modeling capability can be further enhanced while maintaining a controllable amount of model parameters. Extensive experiments on two SSC benchmarks, NYU-Depth-v2 and NYUCAD, show the superior performance of the proposed method. Our code is available at https://waterljwant.github.io/SSC/

cs.CV↗

arXiv2020

DPGN: Distribution Propagation Graph Network for Few-shot Learning

Most graph-network-based meta-learning approaches model instance-level relation of examples. We extend this idea further to explicitly model the distribution-level relation of one example to all other examples in a 1-vs-N manner. We propose a novel approach named distribution propagation graph network (DPGN) for few-shot learning. It conveys both the distribution-level relations and instance-level relations in each few-shot learning task. To combine the distribution-level relations and instance-level relations for all examples, we construct a dual complete graph network which consists of a point graph and a distribution graph with each node standing for an example. Equipped with dual graph architecture, DPGN propagates label information from labeled examples to unlabeled examples within several update generations. In extensive experiments on few-shot learning benchmarks, DPGN outperforms state-of-the-art results by a large margin in 5% $\sim$ 12% under supervised setting and 7% $\sim$ 13% under semi-supervised setting. Code will be released.

cs.CV↗

arXiv2020

Spin-Triplet Excitonic Insulator: The Case of Graphone

While various excitonic insulators have been studied in the literature, due to the perceived too-small spin splitting, spin-triplet excitonic insulator is rare. In two-dimensional systems such as a graphone, however, it is possible, as revealed by first-principles calculations coupled with Bethe-Salpeter equation. The critical temperature, given by an effective Hamiltonian, is 11.5 K. While detecting excitonic insulators is still a daunting challenge, the condensation of triplet excitons will result in spin superfluidity, which can be directly measured by a transport experiment. Nonlocal dielectric screening also leads to an unexpected phenomenon, namely, an indirect-to-direct transition crossover between single-particle band and exciton dispersion in graphone, which offers yet another test by experiment.

cond-mat.mtrl-sci↗

arXiv2020

On the relation between relative rigid and support tilting

Let B be an extriangulated category with enough projectives and enough injectives. Let C be a fully rigid subcategory of B which admits a twin cotorsion pair ((C,K),(K,D)). The quotient category B/K is abelian, we assume that it is hereditary and has finite length. In this article, we study the relation between support tilting subcategories of B/K and maximal relative rigid subcategories of B. More precisely, we show that the image of any cluster tilting subcategory of B is support tilting in B/K and any support tilting subcategory in B/K can be lifted to a unique relative maximal rigid subcategory in B. We also give a bijection between these two classes of subcategories if C is generated by an object.

math.RT↗