arXiv subjects

Ye Yuan

Publications and source records attributed to Ye Yuan.

At least 307 records · Page 17Linked to original sources

arXiv2019

ABD-Net: Attentive but Diverse Person Re-Identification

Attention mechanism has been shown to be effective for person re-identification (Re-ID). However, the learned attentive feature embeddings which are often not naturally diverse nor uncorrelated, will compromise the retrieval performance based on the Euclidean distance. We advocate that enforcing diversity could greatly complement the power of attention. To this end, we propose an Attentive but Diverse Network (ABD-Net), which seamlessly integrates attention modules and diversity regularization throughout the entire network, to learn features that are representative, robust, and more discriminative. Specifically, we introduce a pair of complementary attention modules, focusing on channel aggregation and position awareness, respectively. Furthermore, a new efficient form of orthogonality constraint is derived to enforce orthogonality on both hidden activations and weights. Through careful ablation studies, we verify that the proposed attentive and diverse terms each contributes to the performance gains of ABD-Net. On three popular benchmarks, ABD-Net consistently outperforms existing state-of-the-art methods.

cs.CV↗

arXiv2019

Ego-Pose Estimation and Forecasting as Real-Time PD Control

We propose the use of a proportional-derivative (PD) control based policy learned via reinforcement learning (RL) to estimate and forecast 3D human pose from egocentric videos. The method learns directly from unsegmented egocentric videos and motion capture data consisting of various complex human motions (e.g., crouching, hopping, bending, and motion transitions). We propose a video-conditioned recurrent control technique to forecast physically-valid and stable future motions of arbitrary length. We also introduce a value function based fail-safe mechanism which enables our method to run as a single pass algorithm over the video data. Experiments with both controlled and in-the-wild data show that our approach outperforms previous art in both quantitative metrics and visual quality of the motions, and is also robust enough to transfer directly to real-world scenarios. Additionally, our time analysis shows that the combined use of our pose estimation and forecasting can run at 30 FPS, making it suitable for real-time applications.

cs.CV↗

arXiv2019

Hole compensation effect in III-Mn-V dilute ferromagnetic semiconductors

A systematic study of hole compensation effect on magnetic properties, which is controlled by defect compensation through ion irradiation, in (Ga,Mn)As, (In,Mn)As and (Ga,Mn)P is presented in this work. In all materials, both Curie temperature and magnetization decrease upon increasing the hole compensation, confirming the description of hole mediated ferromagnetism according to the p-d Zener model. The material dependence of Curie temperature and magnetization versus hole compensation reveals that the manipulation of magnetic properties in III-Mn-V dilute ferromagnetic semiconductors by ion irradiation is strongly influenced by the energy level location of the produced defect relative to the band edges in semiconductors.

cond-mat.mtrl-sci↗

arXiv2019

Superconductivity in single-crystalline, aluminum- and gallium-hyperdoped germanium

Superconductivity in group IV semiconductors is desired for hybrid devices combining both semiconducting and superconducting properties. Following boron doped diamond and Si, superconductivity has been observed in gallium doped Ge, however the obtained specimen is in polycrystalline form [Herrmannsdörfer et al., Phys. Rev. Lett. 102, 217003 (2009)]. Here, we present superconducting single-crystalline Ge hyperdoped with gallium or aluminium by ion implantation and rear-side flash lamp annealing. The maximum concentration of Al and Ga incorporated into substitutional positions in Ge is eight times higher than the equilibrium solid solubility. This corresponds to a hole concentration above 10^21 cm-3. Using density functional theory in the local density approximation and pseudopotential plane-wave approach, we show that the superconductivity in p-type Ge is phonon-mediated. According to the ab initio calculations the critical superconducting temperature for Al- and Ga-doped Ge is in the range of 0.45 K for 6.25 at.% of dopant concentration being in a qualitative agreement with experimentally obtained values.

cond-mat.mtrl-sci↗

arXiv2019

A Novel GAN-based Fault Diagnosis Approach for Imbalanced Industrial Time Series

This paper proposes a novel fault diagnosis approach based on generative adversarial networks (GAN) for imbalanced industrial time series where normal samples are much larger than failure cases. We combine a well-designed feature extractor with GAN to help train the whole network. Aimed at obtaining data distribution and hidden pattern in both original distinguishing features and latent space, the encoder-decoder-encoder three-sub-network is employed in GAN, based on Deep Convolution Generative Adversarial Networks (DCGAN) but without Tanh activation layer and only trained on normal samples. In order to verify the validity and feasibility of our approach, we test it on rolling bearing data from Case Western Reserve University and further verify it on data collected from our laboratory. The results show that our proposed approach can achieve excellent performance in detecting faulty by outputting much larger evaluation scores.

cs.LG↗

arXiv2019

Wasserstein Distance based Deep Adversarial Transfer Learning for Intelligent Fault Diagnosis

The demand of artificial intelligent adoption for condition-based maintenance strategy is astonishingly increased over the past few years. Intelligent fault diagnosis is one critical topic of maintenance solution for mechanical systems. Deep learning models, such as convolutional neural networks (CNNs), have been successfully applied to fault diagnosis tasks for mechanical systems and achieved promising results. However, for diverse working conditions in the industry, deep learning suffers two difficulties: one is that the well-defined (source domain) and new (target domain) datasets are with different feature distributions; another one is the fact that insufficient or no labelled data in target domain significantly reduce the accuracy of fault diagnosis. As a novel idea, deep transfer learning (DTL) is created to perform learning in the target domain by leveraging information from the relevant source domain. Inspired by Wasserstein distance of optimal transport, in this paper, we propose a novel DTL approach to intelligent fault diagnosis, namely Wasserstein Distance based Deep Transfer Learning (WD-DTL), to learn domain feature representations (generated by a CNN based feature extractor) and to minimize the distributions between the source and target domains through adversarial training. The effectiveness of the proposed WD-DTL is verified through 3 transfer scenarios and 16 transfer fault diagnosis experiments of both unsupervised and supervised (with insufficient labelled data) learning. We also provide a comprehensive analysis of the network visualization of those transfer tasks.

cs.LG↗

arXiv2019

WIDER Face and Pedestrian Challenge 2018: Methods and Results

This paper presents a review of the 2018 WIDER Challenge on Face and Pedestrian. The challenge focuses on the problem of precise localization of human faces and bodies, and accurate association of identities. It comprises of three tracks: (i) WIDER Face which aims at soliciting new approaches to advance the state-of-the-art in face detection, (ii) WIDER Pedestrian which aims to find effective and efficient approaches to address the problem of pedestrian detection in unconstrained environments, and (iii) WIDER Person Search which presents an exciting challenge of searching persons across 192 movies. In total, 73 teams made valid submissions to the challenge tracks. We summarize the winning solutions for all three tracks. and present discussions on open problems and potential research directions in these topics.

cs.CV↗

arXiv2019

A Full Bayesian Approach to Sparse Network Inference using Heterogeneous Datasets

Network inference has been attracting increasing attention in several fields, notably systems biology, control engineering and biomedicine. To develop a therapy, it is essential to understand the connectivity of biochemical units and the internal working mechanisms of the target network. A network is mainly characterized by its topology and internal dynamics. In particular, sparse topology and stable system dynamics are fundamental properties of many real-world networks. In recent years, kernel-based methods have been popular in the system identification community. By incorporating empirical Bayes, this framework, which we call KEB, is able to promote system stability and impose sparse network topology. Nevertheless, KEB may not be ideal for topology detection due to local optima and numerical errors. Here, therefore, we propose an alternative, data-driven, method that is designed to greatly improve inference accuracy, compared with KEB. The proposed method uses dynamical structure functions to describe networks so that the information of unmeasurable nodes is encoded in the model. A powerful numerical sampling method, namely reversible jump Markov chain Monte Carlo (RJMCMC), is applied to explore full Bayesian models effectively. Monte Carlo simulations indicate that our approach produces more accurate networks compared with KEB methods. Furthermore, simulations of a synthetic biological network demonstrate that the performance of the proposed method is superior to that of the state-of-the-art method, namely iCheMA. The implication is that the proposed method can be used in a wide range of applications, such as controller design, machinery fault diagnosis and therapy development.

eess.SY↗

arXiv2018

Spin-Wave Modes in Transition from a Thin Film to a Full Magnonic Crystal

Spin-wave modes are studied under the gradual transition from a flat thin film to a 'full' (one-dimensional) magnonic crystal. For this purpose, the surface of a pre-patterned 36.8 nm thin permalloy film was sequentially ion milled resulting in magnonic hybrid structures, referred to as surface-modulated magnonic crystals, with increasing modulation depth. After each etching step, ferromagnetic resonance measurements were performed yielding the spin-wave resonance modes in backward-volume and Damon-Eshbach geometry. The spin-wave spectra of these hybrid systems reveal an even larger variety of spin-wave states compared to the 'full' magnonic crystal. The measurements are corroborated by quasi-analytical theory and micromagnetic simulations in order to study the changing spin-wave mode character employing spin-wave mode profiles. In backward-volume geometry, a gradual transition from the uniform mode in the film limit to a fundamental mode in the thin part of the magnonic crystal was observed. Equivalently, the first and the second film modes are transform into a center and an edge mode of the thick part of the magnonic crystal. Simple transition rules from the $n^{\mathrm{th}}$ film mode to the $m^{\mathrm{th}}$ mode in the 'full' magnonic crystal are formulated unraveling the complex mode structure particularly in the backward-volume geometry. An analogous analysis was performed in the Damon-Eshbach geometry.

cond-mat.mes-hall↗

arXiv2018

Nematicity of correlated systems driven by anisotropic chemical phase separation

The origin of nematicity, i.e., in-plane rotational symmetry breaking, and in particular the relative role played by spontaneous unidirectional ordering of spin, orbital, or charge degrees of freedom, is a challenging issue of magnetism, unconventional superconductivity, and quantum Hall effect systems, discussed in the context of doped semiconductor systems, such as Ga$_{1-x}$Mn$_x$As, Cu$_x$Bi$_2$Se$_3$, and Ga(Al)As/Al$_x$Ga$_{1-x}$As quantum wells, respectively. Here, guided by our experimental and theoretical results for In$_{1-x}$Fe$_x$As, we demonstrate that spinodal phase separation at the growth surface (that has a lower symmetry than the bulk) can lead to a quenched nematic order of alloy components, which then governs low temperature magnetic and magnetotransport properties, in particular the magnetoresistance anisotropy whose theory for the $C_{2v}$ symmetry group is advanced here. These findings, together with earlier data for Ga$_{1-x}$Mn$_x$As, show under which conditions anisotropic chemical phase separation accounts for the magnitude of transition temperature to a collective phase or merely breaks its rotational symmetry. We address the question to what extent the directional distribution of impurities or alloy components setting in during the growth may account for the observed nematicity in other classes of correlated systems.

cond-mat.mtrl-sci↗

arXiv2018

Breaking the doping limit in silicon by deep impurities

N-type doping in Si by shallow impurities, such as P, As and Sb, exhibits an intrinsic limit due to the Fermi-level pinning via defect complexes at high doping concentrations. Here we demonstrate that doping Si with the chalcogen Te by non-equilibrium processing, a deep double donor, can exceed this limit and yield higher electron concentrations. In contrast to shallow impurities, both the interstitial Te fraction decreases with increasing doping concentration and substitutional Te dimers become the dominant configuration as effective donors, leading to a non-saturating carrier concentration as well as to an insulator-to-metal transition. First-principle calculations reveal that the Te dimers possess the lowest formation energy and donate two electrons per dimer to the conduction band. These results provide novel insight into physics of deep impurities and lead to a possible solution for the ultra-high electron concentration needed in today's Si-based nanoelectronics.

cond-mat.mtrl-sci↗

arXiv2018

Systems Aliasing in Dynamic Network Reconstruction: Issues on Low Sampling Frequencies

Network reconstruction of dynamical continuous-time (CT) systems is motivated by applications in many fields. Due to experimental limitations, especially in biology, data could be sampled at low frequencies, leading to significant challenges in network inference. We introduce the concept of "system aliasing" and characterize the minimal sampling frequency that allows reconstruction of CT systems from low sampled data. A test criterion is also proposed to check whether system aliasing is presented. With no system aliasing, the paper provides an algorithm to reconstruct dynamic network from data in the presence of noise. In addition, when there is system aliasing we perform studies that add additional prior information of the system such as sparsity. This paper opens new directions in modelling of network systems where samples have significant costs. Such tools are essential to process the available data in applications subject to current experimental limitations.

eess.SY↗

arXiv2018

Epitaxial Mn5Ge3 (100) layer on Ge(100) substrates obtained by flash lamp annealing

Mn5Ge3 thin films have been demonstrated as a promising spin-injector material for germanium-based spintronic devices. So far, Mn5Ge3 has been grown epitaxially only on Ge (111) substrates. In this letter we present the growth of epitaxial Mn5Ge3 films on Ge (100) substrates. The Mn5Ge3 film is synthetized via sub-second solid-state reaction between Mn and Ge upon flash lamp annealing for 20 ms at the ambient pressure. The single crystalline Mn5Ge3 is ferromagnetic with a Curie temperature of 283 K. Both the c-axis of hexagonal Mn5Ge3 and the magnetic easy axis are parallel to the Ge (100) surface. The millisecond-range flash epitaxy provides a new avenue for the fabrication of Ge-based spin-injectors fully compatible with CMOS technology.

cond-mat.mtrl-sci↗

arXiv2018

A Comparative Study of Consistent Snapshot Algorithms for Main-Memory Database Systems

In-memory databases (IMDBs) are gaining increasing popularity in big data applications, where clients commit updates intensively. Specifically, it is necessary for IMDBs to have efficient snapshot performance to support certain special applications (e.g., consistent checkpoint, HTAP). Formally, the in-memory consistent snapshot problem refers to taking an in-memory consistent time-in-point snapshot with the constraints that 1) clients can read the latest data items and 2) any data item in the snapshot should not be overwritten. Various snapshot algorithms have been proposed in academia to trade off throughput and latency, but industrial IMDBs such as Redis adhere to the simple fork algorithm. To understand this phenomenon, we conduct comprehensive performance evaluations on mainstream snapshot algorithms. Surprisingly, we observe that the simple fork algorithm indeed outperforms the state-of-the-arts in update-intensive workload scenarios. On this basis, we identify the drawbacks of existing research and propose two lightweight improvements. Extensive evaluations on synthetic data and Redis show that our lightweight improvements yield better performance than fork, the current industrial standard, and the representative snapshot algorithms from academia. Finally, we have opensourced the implementation of all the above snapshot algorithms so that practitioners are able to benchmark the performance of each algorithm and select proper methods for different application scenarios.

cs.DB↗

arXiv2018

Magnetocrystalline anisotropy and exchange probed by high-field anomalous Hall effect in fully-compensated half-metallic Mn2RuxGa thin films

Magnetotransport is investigated in thin films of the half-metallic ferrimagnet Mn$_2$Ru$_x$Ga in pulsed magnetic fields of up to 58 T. A non-vanishing Hall signal is observed over a broad temperature range, spanning the compensation temperature 155 K, where the net magnetic moment is strictly zero, the anomalous Hall conductivity is 6673 $Ω^{-1}.m^{-1}$ and the coercivity exceeds 9 T. Molecular field modelling is used to determine the intra- and inter-sublattice exchange constants and from the spin-flop transition we infer the anisotropy of the electrically active sublattice to be 216 kJ/m$^3$ and predict the magnetic resonances frequencies. Exchange and anisotropy are comparable and hard-axis applied magnetic fields result in a tilting of the magnetic moments from their collinear ground state. Our analysis is applicable to collinear ferrimagnetic half-metal systems.

cond-mat.mtrl-sci↗

arXiv2018

Data-driven Discovery of Cyber-Physical Systems

Cyber-physical systems (CPSs) embed software into the physical world. They appear in a wide range of applications such as smart grids, robotics, intelligent manufacture and medical monitoring. CPSs have proved resistant to modeling due to their intrinsic complexity arising from the combination of physical components and cyber components and the interaction between them. This study proposes a general framework for reverse engineering CPSs directly from data. The method involves the identification of physical systems as well as the inference of transition logic. It has been applied successfully to a number of real-world examples ranging from mechanical and electrical systems to medical applications. The novel framework seeks to enable researchers to make predictions concerning the trajectory of CPSs based on the discovered model. Such information has been proven essential for the assessment of the performance of CPS, the design of failure-proof CPS and the creation of design guidelines for new CPSs.

eess.SY↗

arXiv2018

Extended Infrared Photoresponse in Te-Hyperdoped Si at Room Temperature

Presently, silicon photonics requires photodetectors that are sensitive in a broad infrared range, can operate at room temperature, and are suitable for integration with the existing Si-technology process. Here, we demonstrate strong room-temperature sub-band-gap photoresponse of photodiodes based on Si hyperdoped with tellurium. The epitaxially recrystallized Te-hyperdoped Si layers are developed by ion implantation combined with pulsed-laser melting and incorporate Te-dopant concentrations several orders of magnitude above the solid solubility limit. With increasing Te concentration, the Te-hyperdoped layer changes from insulating to quasi-metallic behavior with a finite conductivity as the temperature tends to zero. The optical absorptance is found to increase monotonically with increasing Te concentration and extends well into the mid-infrared range. Temperature-dependent optoelectronic photoresponse unambiguously demonstrates that the extended infrared photoresponsivity from Te-hyperdoped Si p-n photodiodes is mediated by a Te intermediate band within the upper half of the Si band gap. This work contributes to pave the way toward establishing a Si-based broadband infrared photonic system operating at room temperature.

cond-mat.mtrl-sci↗

arXiv2018

Global Network Prediction from Local Node Dynamics

The study of dynamical systems on networks, describing complex interactive processes, provides insight into how network structure affects global behaviour. Yet many methods for network dynamics fail to cope with large or partially-known networks, a ubiquitous situation in real-world applications. Here we propose a localised method, applicable to a broad class of dynamical models on networks, whereby individual nodes monitor and store the evolution of their own state and use these values to approximate, via a simple computation, their own steady state solution. Hence the nodes predict their own final state without actually reaching it. Furthermore, the localised formulation enables nodes to compute global network metrics without knowledge of the full network structure. The method can be used to compute global rankings in the network from local information; to detect community detection from fast, local transient dynamics; and to identify key nodes that compute global network metrics ahead of others. We illustrate some of the applications of the algorithm by efficiently performing web-page ranking for a large internet network and identifying the dynamic roles of inter-neurons in the C. Elegans neural network. The mathematical formulation is simple, widely applicable and easily scalable to real-world datasets suggesting how local computation can provide an approach to the study of large-scale network dynamics.

physics.soc-ph↗