arXiv Science⌕ Search

arXiv subjects

Yong Liu

Publications and source records attributed to Yong Liu.

At least 973 records · Page 54Linked to original sources
arXiv2017

Doping evolution of the anisotropic upper critical fields in iron-based superconductor Ba$_{1-x}$K$_x$Fe$_2$As$_2$

In-plane resistivity measurements as a function of temperature and magnetic field up to 35~T with precise orientation within the crystallographic $ac-$plane were used to study the upper critical field, $H_{c2}$, of the hole-doped iron-based superconductor Ba$_{1-x}$K$_x$Fe$_2$As$_2$. Compositions of the samples studied were spanning from underdoped $x=$0.17 ($T_c$=12~K) and $x$=0.22 ($T_c$=20~K), both in the coexistence range of stripe magnetism and superconductivity, though optimal doping $x$=0.39 ($T_c$=38.4~K), $x$=0.47 ($T_c$=37.2~K), to overdoped $x$=0.65 ($T_c$=22~K), $x$=0.83 ($T_c$=10~K). We find notable doping asymmetry of the shapes of the anisotropic $H_{c2}(T)$ suggesting important role of paramagnetic limiting effects in $H \parallel a$ configuration in overdoped compositions and multi-band effects in underdoped compositions.

cond-mat.supr-con↗
arXiv2017

Robust Visual SLAM with Point and Line Features

In this paper, we develop a robust efficient visual SLAM system that utilizes heterogeneous point and line features. By leveraging ORB-SLAM [1], the proposed system consists of stereo matching, frame tracking, local mapping, loop detection, and bundle adjustment of both point and line features. In particular, as the main theoretical contributions of this paper, we, for the first time, employ the orthonormal representation as the minimal parameterization to model line features along with point features in visual SLAM and analytically derive the Jacobians of the re-projection errors with respect to the line parameters, which significantly improves the SLAM solution. The proposed SLAM has been extensively tested in both synthetic and real-world experiments whose results demonstrate that the proposed system outperforms the state-of-the-art methods in various scenarios.

cs.CV↗
arXiv2017

Prototype tests for a highly granular scintillator-based hadronic calorimeter

Within the CALICE collaboration, several concepts for the hadronic calorimeter of a future lepton collider detector are studied. After having demonstrated the capabilities of the measurement methods in "physics prototypes", the focus now lies on improving their implementation in "technological prototypes", that are scalable to the full linear collider detector. The Analogue Hadronic Calorimeter (AHCAL) concept is a sampling calorimeter of tungsten or steel absorber plates and plastic scintillator tiles read out by silicon photomultipliers (SiPMs) as active components. The front-end electronics is fully integrated into the active layers of the calorimeter and is designed for minimal power consumption (i.e. power pulsing). The versatile electronics enables the prototype to be equipped with different types of scintillator tiles and SiPMs. In recent beam tests, a prototype with $\sim$3700 channels, equipped with several types of scintillator tiles and SiPMs, was exposed to electron, muon and hadron beams. The experience of these beam tests resulted in an optimal detector design with surface-mounted SiPMs suitable for the automated mass assembly. The proceeding will cover topics including the testbeam measurements with the AHCAL technological prototype, the improved detector design and the ongoing development of a large prototype for hadronic showers.

physics.ins-det↗
arXiv2017

Multipath IP Routing on End Devices: Motivation, Design, and Performance

Most end devices are now equipped with multiple network interfaces. Applications can exploit all available interfaces and benefit from multipath transmission. Recently Multipath TCP (MPTCP) was proposed to implement multipath transmission at the transport layer and has attracted lots of attention from academia and industry. However, MPTCP only supports TCP-based applications and its multipath routing flexibility is limited. In this paper, we investigate the possibility of orchestrating multipath transmission from the network layer of end devices, and develop a Multipath IP (MPIP) design consisting of signaling, session and path management, multipath routing, and NAT traversal. We implement MPIP in Linux and Android kernels. Through controlled lab experiments and Internet experiments, we demonstrate that MPIP can effectively achieve multipath gains at the network layer. It not only supports the legacy TCP and UDP protocols, but also works seamlessly with MPTCP. By facilitating user-defined customized routing, MPIP can route traffic from competing applications in a coordinated fashion to maximize the aggregate user Quality-of-Experience.

cs.NI↗
arXiv2017

Classifying space of subcategories and its application

For a collection of subcategories satisfying a fixed set of conditions, for example thick subcategories of a triangulated category, we define a topological space called classifying space of subcategories. We show that this space classifies various prime subcategories in the sense that they bijectively correspond to the closed subsets of the classifying space. Many well-known results of subcategory classification fit into this framework. An example which cannot be classified by a topological space in the above sense is also given.

math.CT↗
arXiv2017

Robust Cost-Sensitive Learning for Recommendation with Implicit Feedback

Recommendation is the task of improving customer experience through personalized recommendation based on users' past feedback. In this paper, we investigate the most common scenario: the user-item (U-I) matrix of implicit feedback. Even though many recommendation approaches are designed based on implicit feedback, they attempt to project the U-I matrix into a low-rank latent space, which is a strict restriction that rarely holds in practice. In addition, although misclassification costs from imbalanced classes are significantly different, few methods take the cost of classification error into account. To address aforementioned issues, we propose a robust framework by decomposing the U-I matrix into two components: (1) a low-rank matrix that captures the common preference, and (2) a sparse matrix that detects the user-specific preference of individuals. A cost-sensitive learning model is embedded into the framework. Specifically, this model exploits different costs in the loss function for the observed and unobserved instances. We show that the resulting non-smooth convex objective can be optimized efficiently by an accelerated projected gradient method with closed-form solutions. Morever, the proposed algorithm can be scaled up to large-sized datasets after a relaxation. The theoretical result shows that even with a small fraction of 1's in the U-I matrix $M\in\mathbb{R}^{n\times m}$, the cost-sensitive error of the proposed model is upper bounded by $O(\fracα{\sqrt{mn}})$, where $α$ is a bias over imbalanced classes. Finally, empirical experiments are extensively carried out to evaluate the effectiveness of our proposed algorithm. Encouraging experimental results show that our algorithm outperforms several state-of-the-art algorithms on benchmark recommendation datasets.

cs.LG↗
arXiv2017

On one phase free boundary problem in $\mathbb{R}^{N}$

We construct a smooth axially symmetric solution to the classical one phase free boundary problem in $\mathbb{R}^{N}$. Its free boundary is of \textquotedblleft catenoid\textquotedblright\ type. This is a higher dimensional analogy of the Hauswirth-Helein-Pacard solution in $\mathbb{R}% ^{2}$ (\cite{Pacard}). The existence of such solution is conjectured in \cite [Remark 2.4]{Pacard}.

math.AP↗
arXiv2017

Highly Granular Calorimeters: Technologies and Results

The CALICE collaboration is developing highly granular calorimeters for experiments at a future lepton collider primarily to establish technologies for particle flow event reconstruction. These technologies also find applications elsewhere, such as detector upgrades for the LHC. Meanwhile, the large data sets collected in an extensive series of beam tests have enabled detailed studies of the properties of hadronic showers in calorimeter systems, resulting in improved simulation models and development of sophisticated reconstruction techniques. In this proceeding, highlights are included from studies of the structure of hadronic showers and results on reconstruction techniques for imaging calorimetry. In addition, current R&D activities within CALICE are summarized, focusing on technological prototypes that address challenges from full detector system integration and production techniques amenable to mass production for electromagnetic and hadronic calorimeters based on silicon, scintillator, and gas techniques.

physics.ins-det↗
arXiv2017

On a free boundary problem and minimal surfaces

From minimal surfaces such as Simons' cone and catenoids, using refined Lyapunov-Schmidt reduction method, we construct new solutions for a free boundary problem whose free boundary has two components. In dimension $8$, using variational arguments, we also obtain solutions which are global minimizers of the corresponding energy functional. This shows that Savin's theorem is optimal.

math.AP↗
arXiv2017

Nondegeneracy of the traveling lump solution to the $2+1$ Toda lattice

We consider the $2+1$ Toda system \[ \frac{1}{4}Δq_{n}=e^{q_{n-1}-q_{n}}-e^{q_{n}-q_{n+1}}\text{ in }\mathbb{R}^{2},\ n\in\mathbb{Z}. \] It has a traveling wave type solution $\left\{ Q_{n}\right\} $ satisfying $Q_{n+1}(x,y)=Q_{n}(x+\frac{1}{2\sqrt{2}},y)$, and is explicitly given by \[ Q_{n}\left( x,y\right) =\ln\frac{\frac{1}{4}+\left( n-1+2\sqrt{2}x\right) ^{2}+4y^{2}}{\frac{1}{4}+\left( n+2\sqrt{2}x\right) ^{2}+4y^{2}}. \] In this paper we prove that \{$Q_{n}$\} is nondegenerate.

math.AP↗
arXiv2017

Large Margin Object Tracking with Circulant Feature Maps

Structured output support vector machine (SVM) based tracking algorithms have shown favorable performance recently. Nonetheless, the time-consuming candidate sampling and complex optimization limit their real-time applications. In this paper, we propose a novel large margin object tracking method which absorbs the strong discriminative ability from structured output SVM and speeds up by the correlation filter algorithm significantly. Secondly, a multimodal target detection technique is proposed to improve the target localization precision and prevent model drift introduced by similar objects or background noise. Thirdly, we exploit the feedback from high-confidence tracking results to avoid the model corruption problem. We implement two versions of the proposed tracker with the representations from both conventional hand-crafted and deep convolution neural networks (CNNs) based features to validate the strong compatibility of the algorithm. The experimental results demonstrate that the proposed tracker performs superiorly against several state-of-the-art algorithms on the challenging benchmark sequences while runs at speed in excess of 80 frames per second. The source code and experimental results will be made publicly available.

cs.CV↗
arXiv2017

Real-time 3D Human Tracking for Mobile Robots with Multisensors

Acquiring the accurate 3-D position of a target person around a robot provides fundamental and valuable information that is applicable to a wide range of robotic tasks, including home service, navigation and entertainment. This paper presents a real-time robotic 3-D human tracking system which combines a monocular camera with an ultrasonic sensor by the extended Kalman filter (EKF). The proposed system consists of three sub-modules: monocular camera sensor tracking model, ultrasonic sensor tracking model and multi-sensor fusion. An improved visual tracking algorithm is presented to provide partial location estimation (2-D). The algorithm is designed to overcome severe occlusions, scale variation, target missing and achieve robust re-detection. The scale accuracy is further enhanced by the estimated 3-D information. An ultrasonic sensor array is employed to provide the range information from the target person to the robot and Gaussian Process Regression is used for partial location estimation (2-D). EKF is adopted to sequentially process multiple, heterogeneous measurements arriving in an asynchronous order from the vision sensor and the ultrasonic sensor separately. In the experiments, the proposed tracking system is tested in both simulation platform and actual mobile robot for various indoor and outdoor scenes. The experimental results show the superior performance of the 3-D tracking system in terms of both the accuracy and robustness.

cs.RO↗
arXiv2017

WiLiTV: A Low-Cost Wireless Framework for Live TV Services

With the evolution of HDTV and Ultra HDTV, the bandwidth requirement for IP-based TV content is rapidly increasing. Consumers demand uninterrupted service with a high Quality of Experience (QoE). Service providers are constantly trying to differentiate themselves by innovating new ways of distributing content more efficiently with lower cost and higher penetration. In this work, we propose a cost-efficient wireless framework (WiLiTV) for delivering live TV services, consisting of a mix of wireless access technologies (e.g. Satellite, WiFi and LTE overlay links). In the proposed architecture, live TV content is injected into the network at a few residential locations using satellite dishes. The content is then further distributed to other homes using a house-to-house WiFi network or via an overlay LTE network. Our problem is to construct an optimal TV distribution network with the minimum number of satellite injection points, while preserving the highest QoE, for different neighborhood densities. We evaluate the framework using realistic time-varying demand patterns and a diverse set of home location data. Our study demonstrates that the architecture requires 75 - 90% fewer satellite injection points, compared to traditional architectures. Furthermore, we show that most cost savings can be obtained using simple and practical relay routing solutions.

cs.MM↗
arXiv2016

A classification of nullity classes in the derived category of a ring

For a commutative Noetherian ring $R$ with finite Krull dimension, we study the nullity classes in $D^c_{fg}(R)$, the full triangulated subcategory $D^c_{fg}(R)$ of the derived category $D(R)$ consisting of objects which can be represented by cofibrant objects with each degree finitely generated. In the light of perversity functions over the prime spectrum $\mathrm{Spec} R$, we prove that there is a complete invariant of nullity classes thus that of aisles (or equivalently, $t$-structures) in $D^c_{fg}(R)$.

math.CT↗
arXiv2016

Parse Geometry from a Line: Monocular Depth Estimation with Partial Laser Observation

Many standard robotic platforms are equipped with at least a fixed 2D laser range finder and a monocular camera. Although those platforms do not have sensors for 3D depth sensing capability, knowledge of depth is an essential part in many robotics activities. Therefore, recently, there is an increasing interest in depth estimation using monocular images. As this task is inherently ambiguous, the data-driven estimated depth might be unreliable in robotics applications. In this paper, we have attempted to improve the precision of monocular depth estimation by introducing 2D planar observation from the remaining laser range finder without extra cost. Specifically, we construct a dense reference map from the sparse laser range data, redefining the depth estimation task as estimating the distance between the real and the reference depth. To solve the problem, we construct a novel residual of residual neural network, and tightly combine the classification and regression losses for continuous depth estimation. Experimental results suggest that our method achieves considerable promotion compared to the state-of-the-art methods on both NYUD2 and KITTI, validating the effectiveness of our method on leveraging the additional sensory information. We further demonstrate the potential usage of our method in obstacle avoidance where our methodology provides comprehensive depth information compared to the solution using monocular camera or 2D laser range finder alone.

cs.CV↗
arXiv2016

RISAS: A Novel Rotation, Illumination, Scale Invariant Appearance and Shape Feature

This paper presents a novel appearance and shape feature, RISAS, which is robust to viewpoint, illumination, scale and rotation variations. RISAS consists of a keypoint detector and a feature descriptor both of which utilise texture and geometric information present in the appearance and shape channels. A novel response function based on the surface normals is used in combination with the Harris corner detector for selecting keypoints in the scene. A strategy that uses the depth information for scale estimation and background elimination is proposed to select the neighbourhood around the keypoints in order to build precise invariant descriptors. Proposed descriptor relies on the ordering of both grayscale intensity and shape information in the neighbourhood. Comprehensive experiments which confirm the effectiveness of the proposed RGB-D feature when compared with CSHOT and LOIND are presented. Furthermore, we highlight the utility of incorporating texture and shape information in the design of both the detector and the descriptor by demonstrating the enhanced performance of CSHOT and LOIND when combined with RISAS detector.

cs.RO↗
arXiv2016

Improved Recurrent Neural Networks for Session-based Recommendations

Recurrent neural networks (RNNs) were recently proposed for the session-based recommendation task. The models showed promising improvements over traditional recommendation approaches. In this work, we further study RNN-based models for session-based recommendations. We propose the application of two techniques to improve model performance, namely, data augmentation, and a method to account for shifts in the input data distribution. We also empirically study the use of generalised distillation, and a novel alternative model that directly predicts item embeddings. Experiments on the RecSys Challenge 2015 dataset demonstrate relative improvements of 12.8% and 14.8% over previously reported results on the Recall@20 and Mean Reciprocal Rank@20 metrics respectively.

cs.LG↗