arXiv subjects

Fan Wu

Publications and source records attributed to Fan Wu.

At least 379 records · Page 21Linked to original sources

arXiv2018

A Hybrid Dynamic-regenerative Damping Scheme for Energy Regeneration in Variable Impedance Actuators

Increasing research efforts have been made to improve the energy efficiency of variable impedance actuators (VIAs) through reduction of energy consumption. However, the harvesting of dissipated energy in such systems remains underexplored. This study proposes a novel variable damping module design enabling energy regeneration in VIAs by exploiting the regenerative braking effect of DC motors. The proposed damping module uses four switches to combine regenerative and dynamic braking, in a hybrid approach that enables energy regeneration without reduction in the range of damping achievable. Numerical simulations and a physical experiment are presented in which the proposed module shows an optimal trade-off between task performance and energy efficiency.

eess.SY↗

arXiv2018

Darwinian Data Structure Selection

Data structure selection and tuning is laborious but can vastly improve an application's performance and memory footprint. Some data structures share a common interface and enjoy multiple implementations. We call them Darwinian Data Structures (DDS), since we can subject their implementations to survival of the fittest. We introduce ARTEMIS a multi-objective, cloud-based search-based optimisation framework that automatically finds optimal, tuned DDS modulo a test suite, then changes an application to use that DDS. ARTEMIS achieves substantial performance improvements for \emph{every} project in $5$ Java projects from DaCapo benchmark, $8$ popular projects and $30$ uniformly sampled projects from GitHub. For execution time, CPU usage, and memory consumption, ARTEMIS finds at least one solution that improves \emph{all} measures for $86\%$ ($37/43$) of the projects. The median improvement across the best solutions is $4.8\%$, $10.1\%$, $5.1\%$ for runtime, memory and CPU usage. These aggregate results understate ARTEMIS's potential impact. Some of the benchmarks it improves are libraries or utility functions. Two examples are gson, a ubiquitous Java serialization framework, and xalan, Apache's XML transformation tool. ARTEMIS improves gson by $16.5$\%, $1\%$ and $2.2\%$ for memory, runtime, and CPU; ARTEMIS improves xalan's memory consumption by $23.5$\%. \emph{Every} client of these projects will benefit from these performance improvements.

cs.SE↗

arXiv2018

Global Semantic Consistency for Zero-Shot Learning

In image recognition, there are many cases where training samples cannot cover all target classes. Zero-shot learning (ZSL) utilizes the class semantic information to classify samples of the unseen categories that have no corresponding samples contained in the training set. In this paper, we propose an end-to-end framework, called Global Semantic Consistency Network (GSC-Net for short), which makes complete use of the semantic information of both seen and unseen classes, to support effective zero-shot learning. We also adopt a soft label embedding loss to further exploit the semantic relationships among classes. To adapt GSC-Net to a more practical setting, Generalized Zero-shot Learning (GZSL), we introduce a parametric novelty detection mechanism. Our approach achieves the state-of-the-art performance on both ZSL and GZSL tasks over three visual attribute datasets, which validates the effectiveness and advantage of the proposed framework.

cs.CV↗

arXiv2018

Nonlinear resonances generate large-scale vortices in the phase space in plasma systems

It is well-known that the resonance phenomena can destroy the adiabatic invariance and cause chaos and mixing. In the present paper we show that the nonlinear wave-particle resonant interaction may cause the emergence of large-scale coherent structures in the phase space. The combined action of the drift due to nonlinear scattering on resonance and trapping (capture) into resonance create a vortex-like structure, where the areas of particle acceleration and deceleration are macroscopically separated. At the same time, nonlinear scattering also creates a diffusion that causes mixing and uniformization in around the vortex.

physics.plasm-ph↗

arXiv2018

ERA: Towards Privacy Preservation and Verifiability for Online Ad Exchanges

Ad exchanges are kind of the most popular online advertising marketplaces for trading ad spaces over the Internet. Ad exchanges run auctions to sell diverse ad spaces on the publishers' web-pages to advertisers, who want to display ads on ad spaces. However, the parties in an ad auction cannot verify whether the auction is carried out correctly or not. Furthermore, the advertisers are usually unwilling to reveal their sensitive bids and identities. In this paper, we jointly consider the auction verifiability and the advertisers' privacy preservation, and thus propose ERA, which is an Efficient, pRivacy-preserving, and verifiAble online auction mechanism for ad exchanges. ERA exploits an order preserving encryption scheme to guarantee privacy preservation, and achieves verifiability by constructing a novel protocol of privacy preserving integer comparison, which is built on the Paillier homomorphic encryption scheme. We extensively evaluate the performance of ERA, and our evaluation results show that ERA satisfies several desirable properties with low computation, communication, and storage overheads, so ERA can be easily deployed in today's ad exchanges.

cs.GT↗

arXiv2018

Joint 3D Face Reconstruction and Dense Alignment with Position Map Regression Network

We propose a straightforward method that simultaneously reconstructs the 3D facial structure and provides dense alignment. To achieve this, we design a 2D representation called UV position map which records the 3D shape of a complete face in UV space, then train a simple Convolutional Neural Network to regress it from a single 2D image. We also integrate a weight mask into the loss function during training to improve the performance of the network. Our method does not rely on any prior face model, and can reconstruct full facial geometry along with semantic meaning. Meanwhile, our network is very light-weighted and spends only 9.8ms to process an image, which is extremely faster than previous works. Experiments on multiple challenging datasets show that our method surpasses other state-of-the-art methods on both reconstruction and alignment tasks by a large margin.

cs.CV↗

arXiv2018

Tuning anti-Klein to Klein tunneling in bilayer graphene

We show that in gapped bilayer graphene, quasiparticle tunneling and the corresponding Berry phase can be controlled such that it exhibits features of single layer graphene such as Klein tunneling. The Berry phase is detected by a high-quality Fabry-Pérot interferometer based on bilayer graphene. By raising the Fermi energy of the charge carriers, we find that the Berry phase can be continuously tuned from $2π$ down to $0.68π$ in gapped bilayer graphene, in contrast to the constant Berry phase of $2π$ in pristine bilayer graphene. Particularly, we observe a Berry phase of $π$, the standard value for single layer graphene. As the Berry phase decreases, the corresponding transmission probability of charge carriers at normal incidence clearly demonstrates a transition from anti-Klein tunneling to nearly perfect Klein tunneling.

cond-mat.mes-hall↗

arXiv2017

Quantum phase space with a basis of Wannier functions

A quantum phase space with Wannier basis is constructed: (i) classical phase space is divided into Planck cells; (ii) a complete set of Wannier functions are constructed with the combination of Kohn's method and Löwdin method such that each Wannier function is localized at a Planck cell. With these Wannier functions one can map a wave function unitarily onto phase space. Various examples are used to illustrate our method and compare it to Wigner function. The advantage of our method is that it can smooth out the oscillations in wave functions without losing any information and is potentially a better tool in studying quantum-classical correspondence. In addition, we point out that our method can be used for time-frequency analysis of signals.

quant-ph↗

arXiv2017

An End-to-End Approach to Natural Language Object Retrieval via Context-Aware Deep Reinforcement Learning

We propose an end-to-end approach to the natural language object retrieval task, which localizes an object within an image according to a natural language description, i.e., referring expression. Previous works divide this problem into two independent stages: first, compute region proposals from the image without the exploration of the language description; second, score the object proposals with regard to the referring expression and choose the top-ranked proposals. The object proposals are generated independently from the referring expression, which makes the proposal generation redundant and even irrelevant to the referred object. In this work, we train an agent with deep reinforcement learning, which learns to move and reshape a bounding box to localize the object according to the referring expression. We incorporate both the spatial and temporal context information into the training procedure. By simultaneously exploiting local visual information, the spatial and temporal context and the referring language a priori, the agent selects an appropriate action to take at each time. A special action is defined to indicate when the agent finds the referred object, and terminate the procedure. We evaluate our model on various datasets, and our algorithm significantly outperforms the compared algorithms. Notably, the accuracy improvement of our method over the recent method GroundeR and SCRC on the ReferItGame dataset are 7.67% and 18.25%, respectively.

cs.CV↗

arXiv2017

Tailoring supercurrent confinement in graphene bilayer weak links

The Josephson effect is one of the most studied macroscopic quantum phenomena in condensed matter physics and has been an essential part of the quantum technologies development over the last decades. It is already used in many applications such as magnetometry, metrology, quantum computing, detectors or electronic refrigeration. However, developing devices in which the induced superconductivity can be monitored, both spatially and in its magnitude, remains a serious challenge. In this work, we have used local gates to control confinement, amplitude and density profile of the supercurrent induced in one-dimensional nanoscale constrictions, defined in bilayer graphene-hexagonal boron nitride van der Waals heterostructures. The combination of resistance gate maps, out-of-equilibrium transport, magnetic interferometry measurements, analytical and numerical modelling enables us to explore highly tunable superconducting weak links. Our study opens the path way to design more complex superconducting circuits based on this principle such as electronic interferometers or transition-edge sensors.

cond-mat.mes-hall↗

arXiv2017

Sustainable Incentives for Mobile Crowdsensing: Auctions, Lotteries, and Trust and Reputation Systems

Proper incentive mechanisms are critical for mobile crowdsensing systems to motivate people to actively and persistently participate. This article provides an exposition of design principles of six incentive mechanisms, drawing special attention to the sustainability issue. We cover three primary classes of incentive mechanisms: auctions, lotteries, and trust and reputation systems, as well as three other frameworks of promising potential: bargaining games, contract theory, and market-driven mechanisms.

cs.GT↗

arXiv2017

Characterizing departure delays of flights in passenger aviation network of United States

Flight delay happens every day in airports all over the world. However, systemic investigation in large scales remains a challenge. We collect primary data of domestic departure records from Bureau of Transportation Statistics of United States, and do empirical statistics with them in form of complementary cumulative distributions functions (CCDFs) and transmission function of the delays. Fourteen main airlines are characterized by two types of CCDFs: shifted power-law and exponentially truncated shifted power-law. By setting up two phenomenological models based on mean-field approximation in temporal regime, we convert effect from other delay factors into a propagation one. Three parameters meaningful in measuring airlines emerge as universal metrics. Moreover, method used here could become a novel approach to revealing practical meanings hidden in temporal big data in wide fields.

physics.soc-ph↗

arXiv2017

Translocation time of a polymer chain through an energy gradient nanopore

The translocation time of a polymer chain through an interaction energy gradient nanopore was studied by Monte Carlo simulations and the Fokker-Planck equation with double-absorbing boundary conditions. Both the simulation and calculation revealed three different behaviors for polymer translocation. These behaviors can be explained qualitatively from free-energy landscapes obtained for polymer translocation at different parameters. Results show that the translocation time of a polymer chain through a nanopore can be tuned by suitably designing the interaction energy gradient.

cond-mat.soft↗

arXiv2016

Anisotropic crystallization in solution processed chalcogenide thin film by linearly polarized laser

The low activation energy associated with amorphous chalcogenide structures offers broad tunability of material properties with laser-based or thermal processing. In this paper, we study near-bandgap laser induced anisotropic crystallization in solution processed arsenic sulfide. The modified electronic bandtail states associated with laser irritation lead to a distinctive photoluminescence spectrum, compared to thermally annealed amorphous glass. Laser crystalized materials exhibit a periodic subwavelength ripples structure in transmission electron microscopy experiments and show polarization dependent photoluminescence. Analysis of the local atomic structure of these materials using laboratory-based X-ray pair distribution function analysis indicates that laser irradiation causes a slight rearrangement at the atomic length scale, with a small percentage of S-S homopolar bonds converting to As-S heteropolar bonds. These results highlight fundamental differences between laser and thermal processing in this important class of materials.

cond-mat.mtrl-sci↗

arXiv2015

Atomic Scale Visualization of Quantum Interference on a Weyl Semimetal Surface by Scanning Tunneling Microscopy/Spectroscopy

Weyl semimetals may open a new era in condensed matter physics, materials science and nanotech after graphene and topological insulators. We report the first atomic scale view of the surface states of a Weyl semimetal (NbP) using scanning tunneling microscopy/spectroscopy. We observe coherent quantum interference patterns that arise from the scattering of quasiparticles near point defects on the surface. The measurements reveal the surface electronic structure both below and above the chemical potential in both real and reciprocal spaces. Moreover, the interference maps uncover the scattering processes of NbP's exotic surface states. Through comparison between experimental data and theoretical calculations, we further discover that the scattering channels are largely restricted by the orbital and/or spin texture of the surface band. The visualization of the scattering processes can help design novel transport effects and electronics on the topological surface of a Weyl semimetal.

cond-mat.mtrl-sci↗

arXiv2015

Three-component Fulde-Ferrell superfluids in a two-dimensional Fermi gas with spin-orbit coupling

We investigate the pairing physics of a three-component spin-orbit coupled Fermi gas in two spatial dimensions. The three atomic hyperfine states of the system are coupled by the recently realized synthetic spin-orbit coupling (SOC), which mixes different hyperfine states into helicity branches in a momentum-dependent manner. As a consequence, the interplay of spin-orbit coupling and the hyperfine-state dependent interactions leads to the emergence of Fulde-Ferrell (FF) pairing states with finite center-of-mass momenta even in the absence of the Fermi-surface asymmetry that is usually mandatory to stabilize an SOC-induced FF state. We show that, for different combinations of spin-dependent interactions, the ground state of the system can either be the conventional Bardeen-Cooper-Schrieffer pairing state with zero center-of-mass momentum or be the FF pairing states. Of particular interest here is the existence of a three-component FF pairing state in which every two out of the three components form FF pairing. We map out the phase diagram of the system and characterize the properties of the three-component FF state, such as the order parameters, the gapless contours and the momentum distributions. Based on these results, we discuss possible experimental detection schemes for the interesting pairing states in the system.

cond-mat.quant-gas↗

arXiv2014

Cross-Layer Scheduling for OFDMA-based Cognitive Radio Systems with Delay and Security Constraints

This paper considers the resource allocation problem in an Orthogonal Frequency Division Multiple Access (OFDMA) based cognitive radio (CR) network, where the CR base station adopts full overlay scheme to transmit both private and open information to multiple users with average delay and power constraints. A stochastic optimization problem is formulated to develop flow control and radio resource allocation in order to maximize the long-term system throughput of open and private information in CR system and ensure the stability of primary system. The corresponding optimal condition for employing full overlay is derived in the context of concurrent transmission of open and private information. An online resource allocation scheme is designed to adapt the transmission of open and private information based on monitoring the status of primary system as well as the channel and queue states in the CR network. The scheme is proven to be asymptotically optimal in solving the stochastic optimization problem without knowing any statistical information. Simulations are provided to verify the analytical results and efficiency of the scheme.

cs.NI↗

arXiv2014

Two-sample extended empirical likelihood for estimating equations

We propose a two-sample extended empirical likelihood for inference on the difference between two p-dimensional parameters defined by estimating equations. The standard two-sample empirical likelihood for the difference is Bartlett correctable but its domain is a bounded subset of the parameter space. We expand its domain through a composite similarity transformation to derive the two-sample extended empirical likelihood which is defined on the full parameter space. The extended empirical likelihood has the same asymptotic distribution as the standard one and can also achieve the second order accuracy of the Bartlett correction. We include two applications to illustrate the use of two-sample empirical likelihood methods and to demonstrate the superior coverage accuracy of the extended empirical likelihood confidence regions.

math.ST↗