arXiv subjects

Wei Zhao

Publications and source records attributed to Wei Zhao.

At least 379 records · Page 21Linked to original sources

arXiv2018

Fast Dynamic Routing Based on Weighted Kernel Density Estimation

Capsules as well as dynamic routing between them are most recently proposed structures for deep neural networks. A capsule groups data into vectors or matrices as poses rather than conventional scalars to represent specific properties of target instance. Besides of pose, a capsule should be attached with a probability (often denoted as activation) for its presence. The dynamic routing helps capsules achieve more generalization capacity with many fewer model parameters. However, the bottleneck that prevents widespread applications of capsule is the expense of computation during routing. To address this problem, we generalize existing routing methods within the framework of weighted kernel density estimation, and propose two fast routing methods with different optimization strategies. Our methods prompt the time efficiency of routing by nearly 40\% with negligible performance degradation. By stacking a hybrid of convolutional layers and capsule layers, we construct a network architecture to handle inputs at a resolution of $64\times{64}$ pixels. The proposed models achieve a parallel performance with other leading methods in multiple benchmarks.

cs.CV↗

arXiv2018

Multi-Perspective Context Aggregation for Semi-supervised Cloze-style Reading Comprehension

Cloze-style reading comprehension has been a popular task for measuring the progress of natural language understanding in recent years. In this paper, we design a novel multi-perspective framework, which can be seen as the joint training of heterogeneous experts and aggregate context information from different perspectives. Each perspective is modeled by a simple aggregation module. The outputs of multiple aggregation modules are fed into a one-timestep pointer network to get the final answer. At the same time, to tackle the problem of insufficient labeled data, we propose an efficient sampling mechanism to automatically generate more training examples by matching the distribution of candidates between labeled and unlabeled data. We conduct our experiments on a recently released cloze-test dataset CLOTH (Xie et al., 2017), which consists of nearly 100k questions designed by professional teachers. Results show that our method achieves new state-of-the-art performance over previous strong baselines.

cs.CL↗

arXiv2018

Real-time Characterization of Transient Double-Soliton Dynamics in a Mode-Locked Fiber Laser

By means of the emerging Dispersive Fourier transformation technique, we captured the pulse-resolved spectral evolution dynamics of the double-soliton (DS) states in a single-walled carbon nanotube based Er-doped fiber laser from the initial fluctuations, monitoring the evolution process up to 10 seconds (corresponding to ~260 million roundtrips) discontinuously. Two distinctly different evolutionary types of DS states have been investigated in detail: splitting from one pulse and forming simultaneously. Relaxation oscillations, beating, transient bound state, spectral broadening and pulse interval dynamics have been observed in the evolution process of the DS operation. Our study will be helpful for the further research of mode-locking operation.

physics.optics↗

arXiv2018

Decentralized Multi-UAV Routing in the Presence of Disturbances

We introduce a decentralized and online path planning technique for a network of unmanned aerial vehicles (UAVs) in the presence of weather disturbances. In our problem setting, the group of UAVs are required to collaboratively visit a set of goals scattered in a 2-dimensional area. Each UAV will have to spend energy to reach these goals, but due to unforeseen disturbances, the required energy may vary over time and does not necessarily conform with the initial forecast and/or pre-computed optimal paths. Thus, we are dealing with two fundamental interrelated problems to find a global optimum at each point of time: (1) energy consumption prediction based on disturbances and, hence, online path replanning, and (2) distributed agreement among all UAVs to divide the remaining unvisited goals based on their positions and energy requirements. Our approach consists of four main components: (i) a distributed algorithm that periodically divides the unvisited goals among all the UAVs based on the current energy requirements of the UAVs, (ii) a local (i.e., UAV-level) $\AStar$-based algorithm that computes the {\em desirable} path for each UAV to reach the nodes assigned to it, (iii) a local PID controller that {\em predicts} the inputs to the UAV (i.e., thrust and moments), and (iv) a planner that computes the required energy and the replanning time period. We validate our proposed solution through a rich set of simulations and show that our approach is significantly more efficient than a best-effort algorithm that directs each idle UAV to visit the closest unvisited goal.

cs.DC↗

arXiv2018

Leveraging Long and Short-term Information in Content-aware Movie Recommendation

Movie recommendation systems provide users with ranked lists of movies based on individual's preferences and constraints. Two types of models are commonly used to generate ranking results: long-term models and session-based models. While long-term models represent the interactions between users and movies that are supposed to change slowly across time, session-based models encode the information of users' interests and changing dynamics of movies' attributes in short terms. In this paper, we propose an LSIC model, leveraging Long and Short-term Information in Content-aware movie recommendation using adversarial training. In the adversarial process, we train a generator as an agent of reinforcement learning which recommends the next movie to a user sequentially. We also train a discriminator which attempts to distinguish the generated list of movies from the real records. The poster information of movies is integrated to further improve the performance of movie recommendation, which is specifically essential when few ratings are available. The experiments demonstrate that the proposed model has robust superiority over competitors and sets the state-of-the-art. We will release the source code of this work after publication.

cs.IR↗

arXiv2018

Yuanfudao at SemEval-2018 Task 11: Three-way Attention and Relational Knowledge for Commonsense Machine Comprehension

This paper describes our system for SemEval-2018 Task 11: Machine Comprehension using Commonsense Knowledge. We use Three-way Attentive Networks (TriAN) to model interactions between the passage, question and answers. To incorporate commonsense knowledge, we augment the input with relation embedding from the graph of general knowledge ConceptNet (Speer et al., 2017). As a result, our system achieves state-of-the-art performance with 83.95% accuracy on the official test data. Code is publicly available at https://github.com/intfloat/commonsense-rc

cs.CL↗

arXiv2018

IoT Security: An End-to-End View and Case Study

In this paper, we present an end-to-end view of IoT security and privacy and a case study. Our contribution is three-fold. First, we present our end-to-end view of an IoT system and this view can guide risk assessment and design of an IoT system. We identify 10 basic IoT functionalities that are related to security and privacy. Based on this view, we systematically present security and privacy requirements in terms of IoT system, software, networking and big data analytics in the cloud. Second, using the end-to-end view of IoT security and privacy, we present a vulnerability analysis of the Edimax IP camera system. We are the first to exploit this system and have identified various attacks that can fully control all the cameras from the manufacturer. Our real-world experiments demonstrate the effectiveness of the discovered attacks and raise the alarms again for the IoT manufacturers. Third, such vulnerabilities found in the exploit of Edimax cameras and our previous exploit of Edimax smartplugs can lead to another wave of Mirai attacks, which can be either botnets or worm attacks. To systematically understand the damage of the Mirai malware, we model propagation of the Mirai and use the simulations to validate the modeling. The work in this paper raises the alarm again for the IoT device manufacturers to better secure their products in order to prevent malware attacks like Mirai.

cs.CR↗

arXiv2018

Security Consideration For Deep Learning-Based Image Forensics

Recently, image forensics community has paied attention to the research on the design of effective algorithms based on deep learning technology and facts proved that combining the domain knowledge of image forensics and deep learning would achieve more robust and better performance than the traditional schemes. Instead of improving it, in this paper, the safety of deep learning based methods in the field of image forensics is taken into account. To the best of our knowledge, this is a first work focusing on this topic. Specifically, we experimentally find that the method using deep learning would fail when adding the slight noise into the images (adversarial images). Furthermore, two kinds of strategys are proposed to enforce security of deep learning-based method. Firstly, an extra penalty term to the loss function is added, which is referred to the 2-norm of the gradient of the loss with respect to the input images, and then an novel training method are adopt to train the model by fusing the normal and adversarial images. Experimental results show that the proposed algorithm can achieve good performance even in the case of adversarial images and provide a safety consideration for deep learning-based image forensics

cs.CV↗

arXiv2018

A unified image reconstruction framework for quantitative dual- and triple-energy CT imaging of material compositions

Many clinical applications depend critically on the accurate differentiation and classification of different types of materials in patient anatomy. This work introduces a unified framework for accurate nonlinear material decomposition and applies it, for the first time, in the concept of triple-energy CT (TECT) for enhanced material differentiation and classification as well as dual-energy CT. The triple-energy data acquisition is implemented at the scales of micro-CT and clinical CT imaging with commercial "TwinBeam" dual-source DECT configuration and a fast kV switching DECT configuration. Material decomposition and quantitative comparison with a photon counting detector and with the presence of a bow-tie filter are also performed. The proposed method provides quantitative material- and energy-selective images examining realistic configurations for both dual- and triple-energy CT measurements. Compared to the polychromatic kV CT images, virtual monochromatic images show superior image quality. For the mouse phantom, quantitative measurements show that the differences between gadodiamide and iodine concentrations obtained using TECT and idealized photon counting CT (PCCT) are smaller than 8 mg/mL and 1 mg/mL, respectively. TECT outperforms DECT for multi-contrast CT imaging and is robust with respect to spectrum estimation. For the thorax phantom, the differences between the concentrations of the contrast map and the corresponding true reference values are smaller than 7 mg/mL for all of the realistic configurations. A unified framework for both dual- and triple-energy CT imaging has been established for the accurate extraction of material compositions; considering currently available commercial DECT configurations. The novel technique is promising to provide an urgently needed solution for several CT-based diagnosis and therapy applications.

physics.med-ph↗

arXiv2018

Deep Reinforcement Learning for Sponsored Search Real-time Bidding

Bidding optimization is one of the most critical problems in online advertising. Sponsored search (SS) auction, due to the randomness of user query behavior and platform nature, usually adopts keyword-level bidding strategies. In contrast, the display advertising (DA), as a relatively simpler scenario for auction, has taken advantage of real-time bidding (RTB) to boost the performance for advertisers. In this paper, we consider the RTB problem in sponsored search auction, named SS-RTB. SS-RTB has a much more complex dynamic environment, due to stochastic user query behavior and more complex bidding policies based on multiple keywords of an ad. Most previous methods for DA cannot be applied. We propose a reinforcement learning (RL) solution for handling the complex dynamic environment. Although some RL methods have been proposed for online advertising, they all fail to address the "environment changing" problem: the state transition probabilities vary between two days. Motivated by the observation that auction sequences of two days share similar transition patterns at a proper aggregation level, we formulate a robust MDP model at hour-aggregation level of the auction data and propose a control-by-model framework for SS-RTB. Rather than generating bid prices directly, we decide a bidding model for impressions of each hour and perform real-time bidding accordingly. We also extend the method to handle the multi-agent problem. We deployed the SS-RTB system in the e-commerce search auction platform of Alibaba. Empirical experiments of offline evaluation and online A/B test demonstrate the effectiveness of our method.

cs.AI↗

arXiv2018

EASM: Efficiency-Aware Switch Migration for Balancing Controller Loads in Software-Defined Networking

Distributed multi-controller deployment is a promising method to achieve a scalable and reliable control plane of Software-Defined Networking (SDN). However, it brings a new challenge for balancing loads on the distributed controllers as the network traffic dynamically changes. The unbalanced load distribution on the controllers will increase response delay for processing flows and reduce the controllers'throughput. Switch migration is an effective approach to solve the problem. However, existing schemes focus only on the load balancing performance but ignore migration efficiency, which may result in high migration costs and unnecessary control overheads. This paper proposes Efficiency-Aware Switch Migration (EASM) to balance the controllers'loads and improve migration efficiency. We introduce load difference matrix and trigger factor to measure load balancing on controllers. We also introduce the migration efficiency problem, which considers load balancing rate and migration cost simultaneously to optimally migrate switches. We propose EASM to efficiently solve to the problem. The simulation results show that EASM outperforms baseline schemes by reducing the controller response time by about 21.9%, improving the controller throughput by 30.4% on average, maintaining good load balancing rate, low migration costs and migration time, when the network scale changes.

cs.NI↗

arXiv2018

Concussion classification via deep learning using whole-brain white matter fiber strains

Developing an accurate and reliable injury predictor is central to the biomechanical studies of traumatic brain injury. State-of-the-art efforts continue to rely on empirical, scalar metrics based on kinematics or model-estimated tissue responses explicitly pre-defined in a specific brain region of interest. They could suffer from loss of information. A single training dataset has also been used to evaluate performance but without cross-validation. In this study, we developed a deep learning approach for concussion classification using implicit features of the entire voxel-wise white matter fiber strains. Using reconstructed American National Football League (NFL) injury cases, leave-one-out cross-validation was employed to objectively compare injury prediction performances against two baseline machine learning classifiers (support vector machine (SVM) and random forest (RF)) and four scalar metrics via univariate logistic regression (Brain Injury Criterion (BrIC), cumulative strain damage measure of the whole brain (CSDM-WB) and the corpus callosum (CSDM-CC), and peak fiber strain in the CC). Feature-based deep learning and machine learning classifiers consistently outperformed all scalar injury metrics across all performance categories in cross-validation (e.g., average accuracy of 0.844 vs. 0.746, and average area under the receiver operating curve (AUC) of 0.873 vs. 0.769, respectively, based on the testing dataset). Nevertheless, deep learning achieved the best cross-validation accuracy, sensitivity, and AUC (e.g., accuracy of 0.862 vs. 0.828 and 0.842 for SVM and RF, respectively). These findings demonstrate the superior performances of deep learning in concussion prediction, and suggest its promise for future applications in biomechanical investigations of traumatic brain injury.

q-bio.QM↗

arXiv2018

Imaging and variability studies of CTA~102 during the 2016 January $γ$-ray flare

The $γ$-ray bright blazar CTA 102 is studied using imaging (new 15 GHz and archival 43 GHz Very Long Baseline Array, VLBA data) and time variable optical flux density, polarization degree and electric vector position angle (EVPA) spanning between 2015 June 1 and 2016 October 1, covering a prominent $γ$-ray flare during 2016 January. The pc-scale jet indicates expansion with oscillatory features upto 17 mas. Component proper motions are in the range 0.04 - 0.33 mas/yr with acceleration upto 1.2 mas followed by a slowing down beyond 1.5 mas. A jet bulk Lorentz factor $\geq$ 17.5, position angle of 128.3 degrees, inclination angle $\leq$ 6.6 degrees and intrinsic half opening angle $\leq$ 1.8 degrees are derived from the VLBA data. These inferences are employed in a helical jet model to infer long term variability in flux density, polarization degree, EVPA and a rotation of the Stokes Q and U parameters. A core distance of $r_{\rm core,43 \ GHz}$ = 22.9 pc, and a magnetic field strength at 1 pc and the core location of 1.57 G and 0.07 G respectively are inferred using the core shift method. The study is useful in the context of estimating jet parameters and in offering clues to distinguish mechanisms responsible for variability over different timescales.

astro-ph.HE↗

arXiv2017

Structure Re-determination and Superconductivity Observation of Bulk 1T MoS2

2H MoS2 has been intensively studied because of layer-dependent electronic structures and novel physical properties. Though the metastable 1T MoS2 with the [MoS6] octahedron was observed from the microscopic area, the true crystal structure of 1T phase has not been determined strictly. Moreover, the true physical properties have not been demonstrated from experiments due to the challenge for the preparation of pure 1T MoS2 crystals. Here, we successfully synthesized the 1T MoS2 single crystals and re-determined the crystal structure of 1T MoS2 from single-crystal X-ray diffraction. 1T MoS2 crystalizes in space group P-3m1 with a cell of a = b = 3.190(3) Å and c = 5.945(6) Å. The individual MoS2 layer consists of MoS6 octahedron sharing edge with each other. More surprisingly, the bulk 1T MoS2 crystals undergo a superconducting transition of Tc = 4 K, which is the first observation of superconductivity in pure 1T MoS2 phase.

cond-mat.supr-con↗

arXiv2017

Closed-Form Error Analysis on RSS-based Indoor Localization Method

Received Signal Strength (RSS) is considered as a promising measurement for indoor positioning. Lots of RSS-based localization methods have been proposed by its convenience and low cost. This paper focuses on two challenging issues in RSS-based localization schemes: finding optimal localization algorithms and knowing what affects the accuracy of these algorithms. Through theoretical and experimental analysis, we present three important results: 1) we prove that the Non-linear Least Square (NLS) method is efficient to solve the RSS-based localization problem, i.e., its localization error is minimal in theory; 2) we provide a closed-form expression of the localization error for the NLS method. Such an expression reveals the key factors that affect the accuracy of the NLS method; 3) we further conduct a new lower bound of localization error, i.e., the best accuracy achieved possibly. The study of this paper shows the inherent limitation of RSS-based localization schemes and provides guidance for achieving optimal accuracy.

cs.NI↗

arXiv2017

A comparison theorem for Finsler submanifolds and its applications

In this paper, we consider the conormal bundle over a submanifold in a Finsler manifold and establish a volume comparison theorem. As an application, we derive a lower estimate for length of closed geodesics in a Finsler manifold. In the reversible case, a lower bound of injective radius is also obtained.

math.DG↗

arXiv2017

Topological Dirac states beyond $π$ orbitals for silicene on SiC(0001) surface

The discovery of intriguing properties related to the Dirac states in graphene has spurred huge interest in exploring its two-dimensional group-IV counterparts, such as silicene, germanene, and stanene. However, these materials have to be obtained via synthesizing on substrates with strong interfacial interactions, which usually destroy their intrinsic $π$($p_z$)-orbital Dirac states. Here we report a theoretical study on the existence of Dirac states arising from the $p_{x,y}$ orbitals instead of $p_z$ orbitals in silicene on 4H-SiC(0001), which survive in spite of the strong interfacial interactions. We also show that the exchange field together with the spin-orbital coupling give rise to a detectable band gap of 1.3 meV. Berry curvature calculations demonstrate the nontrivial topological nature of such Dirac states with a Chern number $C = 2$, presenting the potential of realizing quantum anomalous Hall effect for silicene on SiC(0001). Finally, we construct a minimal effective model to capture the low-energy physics of this system. This finding is expected to be also applicable to germanene and stanene, and imply great application potentials in nanoelectronics.

cond-mat.mes-hall↗

arXiv2017

Improved Hardy and Rellich inequalities on nonreversible Finsler manifolds

In this paper, we study the sharp constants of quantitative Hardy and Rellich inequalities on nonreversible Finsler manifolds equipped with arbitrary measures. In particular, these inequalities can be globally refined by adding remainder terms like the Brezis-Vázquez improvement, if Finsler manifolds are of strictly negative flag curvature, vanishing S-curvature and finite uniformity constant. Furthermore, these results remain valid when Finsler metrics are reversible.

math.DG↗