arXiv subjects

Yan Zhang

Publications and source records attributed to Yan Zhang.

At least 613 records · Page 34Linked to original sources

arXiv2019

Local Temporal Bilinear Pooling for Fine-grained Action Parsing

Fine-grained temporal action parsing is important in many applications, such as daily activity understanding, human motion analysis, surgical robotics and others requiring subtle and precise operations in a long-term period. In this paper we propose a novel bilinear pooling operation, which is used in intermediate layers of a temporal convolutional encoder-decoder net. In contrast to other work, our proposed bilinear pooling is learnable and hence can capture more complex local statistics than the conventional counterpart. In addition, we introduce exact lower-dimension representations of our bilinear forms, so that the dimensionality is reduced with neither information loss nor extra computation. We perform intensive experiments to quantitatively analyze our model and show the superior performances to other state-of-the-art work on various datasets.

cs.CV↗

arXiv2019

Robust Unsupervised Flexible Auto-weighted Local-Coordinate Concept Factorization for Image Clustering

We investigate the high-dimensional data clustering problem by proposing a novel and unsupervised representation learning model called Robust Flexible Auto-weighted Local-coordinate Concept Factorization (RFA-LCF). RFA-LCF integrates the robust flexible CF, robust sparse local-coordinate coding and the adaptive reconstruction weighting learning into a unified model. The adaptive weighting is driven by including the joint manifold preserving constraints on the recovered clean data, basis concepts and new representation. Specifically, our RFA-LCF uses a L2,1-norm based flexible residue to encode the mismatch between clean data and its reconstruction, and also applies the robust adaptive sparse local-coordinate coding to represent the data using a few nearby basis concepts, which can make the factorization more accurate and robust to noise. The robust flexible factorization is also performed in the recovered clean data space for enhancing representations. RFA-LCF also considers preserving the local manifold structures of clean data space, basis concept space and the new coordinate space jointly in an adaptive manner way. Extensive comparisons show that RFA-LCF can deliver enhanced clustering results.

cs.CV↗

arXiv2019

Joint Label Prediction based Semi-Supervised Adaptive Concept Factorization for Robust Data Representation

Constrained Concept Factorization (CCF) yields the enhanced representation ability over CF by incorporating label information as additional constraints, but it cannot classify and group unlabeled data appropriately. Minimizing the difference between the original data and its reconstruction directly can enable CCF to model a small noisy perturbation, but is not robust to gross sparse errors. Besides, CCF cannot preserve the manifold structures in new representation space explicitly, especially in an adaptive manner. In this paper, we propose a joint label prediction based Robust Semi-Supervised Adaptive Concept Factorization (RS2ACF) framework. To obtain robust representation, RS2ACF relaxes the factorization to make it simultaneously stable to small entrywise noise and robust to sparse errors. To enrich prior knowledge to enhance the discrimination, RS2ACF clearly uses class information of labeled data and more importantly propagates it to unlabeled data by jointly learning an explicit label indicator for unlabeled data. By the label indicator, RS2ACF can ensure the unlabeled data of the same predicted label to be mapped into the same class in feature space. Besides, RS2ACF incorporates the joint neighborhood reconstruction error over the new representations and predicted labels of both labeled and unlabeled data, so the manifold structures can be preserved explicitly and adaptively in the representation space and label space at the same time. Owing to the adaptive manner, the tricky process of determining the neighborhood size or kernel width can be avoided. Extensive results on public databases verify that our RS2ACF can deliver state-of-the-art data representation, compared with other related methods.

cs.CV↗

arXiv2019

A Comparison of Methods for Identifying Location Effects in Unreplicated Fractional Factorials in the Presence of Dispersion Effects

Most methods for identifying location effects in unreplicated fractional factorial designs assume homoscedasticity of the response values. However, dispersion effects in the underlying process may create heteroscedasticity in the response values. This heteroscedasticity may go undetected when identification of location effects is pursued. Indeed, methods for identifying dispersion effects typically require first modeling location effects. Therefore, it is imperative to understand how methods for identifying location effects function in the presence of undetected dispersion effects. We used simulation studies to examine the robustness of four different methods for identifying location effects---Box and Meyer (1986), Lenth (1989), Berk and Picard (1991), and Loughin and Noble (1997)---under models with one, two, or three dispersion effects of varying sizes. We found that the first three methods usually performed acceptably with respect to error rates and power, but the Loughin-Noble method lost control of the individual error rate when moderate-to-large dispersion effects were present.

stat.ME↗

arXiv2019

Super-resolution Imaging of the Fluorescent Dipole Assembly with Polarized Structured Illumination Microscopy

Fluorescence polarization microscopy images both the intensity and orientation of fluorescent dipoles, which plays a vital role in studying the molecular structure and dynamics of bio-complex. However, it is difficult to resolve the dipole assemblies on the subcellular structure and their dynamics in living cells with super-resolution. Here we report polarized structured illumination microscopy (pSIM), which decouples the entangled spatial and angular structured illumination through interpreting the dipoles in spatio-angular hyperspace. We demonstrate its application on a series of biological filamentous systems such as cytoskeleton networks and lambda-DNA, and report the dynamics of short actin sliding through myosin-coated surface. Further, pSIM reveals "side-by-side" organization of the actin ring structure in the membrane-associated periodic skeleton in hippocampal neurons. It also images the dipole dynamics of green fluorescent proteins labeled to the microtubules in live U2OS cells. pSIM can be applied directly to a large variety of commercial or home-built SIM systems.

physics.optics↗

arXiv2019

Visualization of Convolutional Neural Networks for Monocular Depth Estimation

Recently, convolutional neural networks (CNNs) have shown great success on the task of monocular depth estimation. A fundamental yet unanswered question is: how CNNs can infer depth from a single image. Toward answering this question, we consider visualization of inference of a CNN by identifying relevant pixels of an input image to depth estimation. We formulate it as an optimization problem of identifying the smallest number of image pixels from which the CNN can estimate a depth map with the minimum difference from the estimate from the entire image. To cope with a difficulty with optimization through a deep CNN, we propose to use another network to predict those relevant image pixels in a forward computation. In our experiments, we first show the effectiveness of this approach, and then apply it to different depth estimation networks on indoor and outdoor scene datasets. The results provide several findings that help exploration of the above question.

cs.CV↗

arXiv2019

Distributed off-Policy Actor-Critic Reinforcement Learning with Policy Consensus

In this paper, we propose a distributed off-policy actor critic method to solve multi-agent reinforcement learning problems. Specifically, we assume that all agents keep local estimates of the global optimal policy parameter and update their local value function estimates independently. Then, we introduce an additional consensus step to let all the agents asymptotically achieve agreement on the global optimal policy function. The convergence analysis of the proposed algorithm is provided and the effectiveness of the proposed algorithm is validated using a distributed resource allocation example. Compared to relevant distributed actor critic methods, here the agents do not share information about their local tasks, but instead they coordinate to estimate the global policy function.

cs.LG↗

arXiv2019

Fractional spectral graph wavelets and their applications

One of the key challenges in the area of signal processing on graphs is to design transforms and dictionaries methods to identify and exploit structure in signals on weighted graphs. In this paper, we first generalize graph Fourier transform (GFT) to graph fractional Fourier transform (GFRFT), which is then used to define a novel transform named spectral graph fractional wavelet transform (SGFRWT), which is a generalized and extended version of spectral graph wavelet transform (SGWT). A fast algorithm for SGFRWT is also derived and implemented based on Fourier series approximation. The potential applications of SGFRWT are also presented.

cs.CV↗

arXiv2019

Understanding the effect of porosity on the polarisation-field response of ferroelectric materials

This paper combines experimental and modelling studies to provide a detailed examination of the influence of porosity volume fraction and morphology on the polarisation-electric field response of ferroelectric materials. The broadening of the electric field distribution and a decrease in the electric field experienced by the ferroelectric ceramic medium due to the presence of low-permittivity pores is examined and its implications on the shape of the hysteresis loop, remnant polarisation and coercive field is discussed. The variation of coercive field with porosity level is seen to be complex and is attributed to two competing mechanisms where at high porosity levels the influence of the broadening of the electric field distribution dominates, while at low porosity levels an increase in the compliance of the matrix is more important. This new approach to understanding these materials enables the seemingly conflicting observations in the existing literature to be clarified and provides an effective approach to interpret the influence of pore fraction and morphology on the polarisation behaviour of ferroelectrics. Such information provides new insights in the interpretation of the physical properties of porous ferroelectric materials to inform future effort in the design of ferroelectric materials for piezoelectric sensor, actuator, energy harvesting, and transducer applications.

cond-mat.mtrl-sci↗

arXiv2019

High piezoelectric sensitivity and hydrostatic figures of merit in unidirectional porous ferroelectric ceramics fabricated by freeze casting

High performance lead zirconate titanate (PZT) ceramics with aligned porosity for sensing applications were fabricated by an ice-templating method. To demonstrate the enhanced properties of these materials and their potential for sensor and hydrophone applications, the piezoelectric voltage constants hydrostatic parameters and AC conductivity as a function of the porosity in directions both parallel and perpendicular to the freezing temperature gradient were studied. As the porosity level was increased, PZT poled parallel to the freezing direction exhibited the highest coefficients, and hydrostatic figures of merit compared to the dense and PZT poled perpendicular to the freezing direction. This work demonstrates that piezoelectric ceramics produced with aligned pores by freeze casting are a promising candidate for a range of sensor applications and the polarisation orientation relative to the freezing direction can be used to tailor the microstructure and optimise sensitivity for sensor and hydrostatic transducer applications.

physics.app-ph↗

arXiv2019

Energy harvesting from coupled bending-twisting oscillations in carbon-fibre reinforced polymer laminates

The energy harvesting capability of resonant harvesting structures, such as piezoelectric cantilever beams, can be improved by utilizing coupled oscillations that generate favourable strain mode distributions. In this work, we present the first demonstration of the use of a laminated carbon fibre reinforced polymer to create cantilever beams that undergo coupled bending-twisting oscillations for energy harvesting applications. Piezoelectric layers that operate in bending and shear mode are attached to the bend-twist coupled beam surface at locations of maximum bending and torsional strains in the first mode of vibration to fully exploit the strain distribution along the beam. Modelling of this new bend-twist harvesting system is presented, which compares favourably with experimental results. It is demonstrated that the variety of bend and torsional modes of the harvesters can be utilized to create a harvester that operates over a wider range of frequencies and such multi-modal device architectures provides a unique approach to tune the frequency response of resonant harvesting systems.

cond-mat.mtrl-sci↗

arXiv2019

Enhanced pyroelectric and piezoelectric properties of PZT with aligned porosity for energy harvesting applications

This paper demonstrates the significant benefits of exploiting highly aligned porosity in piezoelectric and pyroelectric materials for improved energy harvesting performance. Porous lead zirconate (PZT) ceramics with aligned pore channels and varying fractions of porosity were manufactured in a water-based suspension using freeze casting. The aligned porous PZT ceramics were characterized in detail for both piezoelectric and pyroelectric properties and their energy harvesting performance figures of merit were assessed parallel and perpendicular to the freezing direction. As a result of the introduction of porosity into the ceramic microstrucutre, high piezoelectric and pyroelectric harvesting figures of merits were achieved for porous freeze-cast PZT compared to dense PZT due to the reduced permittivity and volume specific heat capacity. Experimental results were compared to parallel and series analytical models with good agreement and the PZT with porosity aligned parallel to the freezing direction exhibited the highest piezoelectric and pyroelectric harvesting response; this was a result of the enhanced interconnectivity of the ferroelectric material along the poling direction and reduced fraction of unpoled material that leads to a higher polarization. A complete thermal energy harvesting system, composed of an parallel-aligned PZT harvester element and an AC/DC converter successfully demonstrated by charging a storage capacitor. The maximum energy density generated by the 60 vol.% porous parallel-connected PZT when subjected to thermal oscillations was 1653 μJ/cm3 respectively, which was 374% higher than that of the dense PZT with an energy density of 446 μJ/cm3. The results are of benefit for the design and manufacture of high performance porous pyroelectric and piezoelectric materials in devices for energy harvesting and sensor applications.

physics.app-ph↗

arXiv2019

Monolithic integration of broadband optical isolators for polarization-diverse silicon photonics

Integrated optical isolators have been a longstanding challenge for photonic integrated circuits (PIC). An ideal integrated optical isolator for PIC should be made by a monolithic process, have a small footprint, exhibit broadband and polarization-diverse operation, and be compatible with multiple materials platforms. Despite significant progress, the optical isolators reported so far do not meet all these requirements. In this article we present monolithically integrated broadband magneto-optical isolators on silicon and silicon nitride (SiN) platforms operating for both TE and TM modes with record high performances, fulfilling all the essential characteristics for PIC applications. In particular, we demonstrate fully-TE broadband isolators by depositing high quality magneto-optical garnet thin films on the sidewalls of Si and SiN waveguides, a critical result for applications in TE-polarized on-chip lasers and amplifiers. This work demonstrates monolithic integration of high performance optical isolators on chip for polarization-diverse silicon photonic systems, enabling new pathways to impart nonreciprocal photonic functionality to a variety of integrated photonic devices.

physics.app-ph↗

arXiv2019

Neural IR Meets Graph Embedding: A Ranking Model for Product Search

Recently, neural models for information retrieval are becoming increasingly popular. They provide effective approaches for product search due to their competitive advantages in semantic matching. However, it is challenging to use graph-based features, though proved very useful in IR literature, in these neural approaches. In this paper, we leverage the recent advances in graph embedding techniques to enable neural retrieval models to exploit graph-structured data for automatic feature extraction. The proposed approach can not only help to overcome the long-tail problem of click-through data, but also incorporate external heterogeneous information to improve search results. Extensive experiments on a real-world e-commerce dataset demonstrate significant improvement achieved by our proposed approach over multiple strong baselines both as an individual retrieval model and as a feature used in learning-to-rank frameworks.

cs.IR↗

arXiv2019

Possible phonon-induced electronic bi-stability in VO$_2$ for ultrafast memory at room temperature

VO$_{2}$ is a model material system which exhibits a metal to insulator transition at 67$^\circ$C. This holds potential for future ultrafast switching in memory devices, but typically requires a purely electronic process to avoid the slow lattice response. The role of lattice vibrations is thus important, but it is not well understood and it has been a long-standing source of controversy. We use a combination of ultrafast spectroscopy and ab initio quantum calculations to unveil the mechanism responsible for the transition. We identify an atypical Peierls vibrational mode which acts as a trigger for the transition. This rules out the long standing paradigm of a purely electronic Mott transition in VO$_{2}$; however, we found a new electron-phonon pathway for a purely reversible electronic transition in a true bi-stable fashion under specific conditions. This transition is very atypical, as it involves purely charge-like excitations and requires only small nuclear displacement. Our findings will prompt the design of future ultrafast electro-resistive non-volatile memory devices.

cond-mat.str-el↗

arXiv2019

UAV Communications for 5G and Beyond: Recent Advances and Future Trends

Providing ubiquitous connectivity to diverse device types is the key challenge for 5G and beyond 5G (B5G). Unmanned aerial vehicles (UAVs) are expected to be an important component of the upcoming wireless networks that can potentially facilitate wireless broadcast and support high rate transmissions. Compared to the communications with fixed infrastructure, UAV has salient attributes, such as flexible deployment, strong line-of-sight (LoS) connection links, and additional design degrees of freedom with the controlled mobility. In this paper, a comprehensive survey on UAV communication towards 5G/B5G wireless networks is presented. We first briefly introduce essential background and the space-air-ground integrated networks, as well as discuss related research challenges faced by the emerging integrated network architecture. We then provide an exhaustive review of various 5G techniques based on UAV platforms, which we categorize by different domains including physical layer, network layer, and joint communication, computing and caching. In addition, a great number of open research problems are outlined and identified as possible future research directions.

cs.NI↗

arXiv2019

Learning Representations of Sets through Optimized Permutations

Representations of sets are challenging to learn because operations on sets should be permutation-invariant. To this end, we propose a Permutation-Optimisation module that learns how to permute a set end-to-end. The permuted set can be further processed to learn a permutation-invariant representation of that set, avoiding a bottleneck in traditional set models. We demonstrate our model's ability to learn permutations and set representations with either explicit or implicit supervision on four datasets, on which we achieve state-of-the-art results: number sorting, image mosaics, classification from image mosaics, and visual question answering.

cs.LG↗

arXiv2018

An Empirical Study towards Understanding How Deep Convolutional Nets Recognize Falls

Detecting unintended falls is essential for ambient intelligence and healthcare of elderly people living alone. In recent years, deep convolutional nets are widely used in human action analysis, based on which a number of fall detection methods have been proposed. Despite their highly effective performances, the behaviors of how the convolutional nets recognize falls are still not clear. In this paper, instead of proposing a novel approach, we perform a systematical empirical study, attempting to investigate the underlying fall recognition process. We propose four tasks to investigate, which involve five types of input modalities, seven net instances and different training samples. The obtained quantitative and qualitative results reveal the patterns that the nets tend to learn, and several factors that can heavily influence the performances on fall recognition. We expect that our conclusions are favorable to proposing better deep learning solutions to fall detection systems.

cs.CV↗