Searcharxiv⌕ Search

arXiv subjects

Qian Zhang

Publications and source records attributed to Qian Zhang.

At least 415 records · Page 23Linked to original sources
arXiv2019

Reinforced Evolutionary Neural Architecture Search

Neural Architecture Search (NAS) is an important yet challenging task in network design due to its high computational consumption. To address this issue, we propose the Reinforced Evolutionary Neural Architecture Search (RE- NAS), which is an evolutionary method with the reinforced mutation for NAS. Our method integrates reinforced mutation into an evolution algorithm for neural architecture exploration, in which a mutation controller is introduced to learn the effects of slight modifications and make mutation actions. The reinforced mutation controller guides the model population to evolve efficiently. Furthermore, as child models can inherit parameters from their parents during evolution, our method requires very limited computational resources. In experiments, we conduct the proposed search method on CIFAR-10 and obtain a powerful network architecture, RENASNet. This architecture achieves a competitive result on CIFAR-10. The explored network architecture is transferable to ImageNet and achieves a new state-of-the-art accuracy, i.e., 75.7% top-1 accuracy with 5.36M parameters on mobile ImageNet. We further test its performance on semantic segmentation with DeepLabv3 on the PASCAL VOC. RENASNet outperforms MobileNet-v1, MobileNet-v2 and NASNet. It achieves 75.83% mIOU without being pre-trained on COCO.

cs.NE↗
arXiv2019

Towards Motion Invariant Authentication for On-Body IoT Devices

As the rapid proliferation of on-body Internet of Things (IoT) devices, their security vulnerabilities have raised serious privacy and safety issues. Traditional efforts to secure these devices against impersonation attacks mainly rely on either dedicated sensors or specified user motions, impeding their wide-scale adoption. This paper transcends these limitations with a general security solution by leveraging ubiquitous wireless chips available in IoT devices. Particularly, representative time and frequency features are first extracted from received signal strengths (RSSs) to characterize radio propagation profiles. Then, an adversarial multi-player network is developed to recognize underlying radio propagation patterns and facilitate on-body device authentication. We prove that at equilibrium, our adversarial model can extract all information about propagation patterns and eliminate any irrelevant information caused by motion variances. We build a prototype of our system using universal software radio peripheral (USRP) devices and conduct extensive experiments with both static and dynamic body motions in typical indoor and outdoor environments. The experimental results show that our system achieves an average authentication accuracy of 90.4%, with a high area under the receiver operating characteristic curve (AUROC) of 0.958 and better generalization performance in comparison with the conventional non-adversarial-based approach.

cs.CR↗
arXiv2019

Simultaneous Spectral-Spatial Feature Selection and Extraction for Hyperspectral Images

In hyperspectral remote sensing data mining, it is important to take into account of both spectral and spatial information, such as the spectral signature, texture feature and morphological property, to improve the performances, e.g., the image classification accuracy. In a feature representation point of view, a nature approach to handle this situation is to concatenate the spectral and spatial features into a single but high dimensional vector and then apply a certain dimension reduction technique directly on that concatenated vector before feed it into the subsequent classifier. However, multiple features from various domains definitely have different physical meanings and statistical properties, and thus such concatenation hasn't efficiently explore the complementary properties among different features, which should benefit for boost the feature discriminability. Furthermore, it is also difficult to interpret the transformed results of the concatenated vector. Consequently, finding a physically meaningful consensus low dimensional feature representation of original multiple features is still a challenging task. In order to address the these issues, we propose a novel feature learning framework, i.e., the simultaneous spectral-spatial feature selection and extraction algorithm, for hyperspectral images spectral-spatial feature representation and classification. Specifically, the proposed method learns a latent low dimensional subspace by projecting the spectral-spatial feature into a common feature space, where the complementary information has been effectively exploited, and simultaneously, only the most significant original features have been transformed. Encouraging experimental results on three public available hyperspectral remote sensing datasets confirm that our proposed method is effective and efficient.

cs.CV↗
arXiv2019

EAT-NAS: Elastic Architecture Transfer for Accelerating Large-scale Neural Architecture Search

Neural architecture search (NAS) methods have been proposed to release human experts from tedious architecture engineering. However, most current methods are constrained in small-scale search due to the issue of computational resources. Meanwhile, directly applying architectures searched on small datasets to large datasets often bears no performance guarantee. This limitation impedes the wide use of NAS on large-scale tasks. To overcome this obstacle, we propose an elastic architecture transfer mechanism for accelerating large-scale neural architecture search (EAT-NAS). In our implementations, architectures are first searched on a small dataset, e.g., CIFAR-10. The best one is chosen as the basic architecture. The search process on the large dataset, e.g., ImageNet, is initialized with the basic architecture as the seed. The large-scale search process is accelerated with the help of the basic architecture. What we propose is not only a NAS method but a mechanism for architecture-level transfer. In our experiments, we obtain two final models EATNet-A and EATNet-B that achieve competitive accuracies, 74.7% and 74.2% on ImageNet, respectively, which also surpass the models searched from scratch on ImageNet under the same settings. For the computational cost, EAT-NAS takes only less than 5 days on 8 TITAN X GPUs, which is significantly less than the computational consumption of the state-of-the-art large-scale NAS methods.

cs.CV↗
arXiv2019

Enhancing Remote Sensing Image Retrieval with Triplet Deep Metric Learning Network

With the rapid growing of remotely sensed imagery data, there is a high demand for effective and efficient image retrieval tools to manage and exploit such data. In this letter, we present a novel content-based remote sensing image retrieval method based on Triplet deep metric learning convolutional neural network (CNN). By constructing a Triplet network with metric learning objective function, we extract the representative features of the images in a semantic space in which images from the same class are close to each other while those from different classes are far apart. In such a semantic space, simple metric measures such as Euclidean distance can be used directly to compare the similarity of images and effectively retrieve images of the same class. We also investigate a supervised and an unsupervised learning methods for reducing the dimensionality of the learned semantic features. We present comprehensive experimental results on two publicly available remote sensing image retrieval datasets and show that our method significantly outperforms state-of-the-art.

cs.CV↗
arXiv2018

Cross-Technology Communications for Heterogeneous IoT Devices Through Artificial Doppler Shifts

Recent years have seen major innovations in developing energy-efficient wireless technologies such as Bluetooth Low Energy (BLE) for Internet of Things (IoT). Despite demonstrating significant benefits in providing low power transmission and massive connectivity, hardly any of these technologies have made it to directly connect to the Internet. Recent advances demonstrate the viability of direct communication among heterogeneous IoT devices with incompatible physical (PHY) layers. These techniques, however, require modifications in transmission power or time, which may affect the media access control (MAC) layer behaviors in legacy networks. In this paper, we argue that the frequency domain can serve as a free side channel with minimal interruptions to legacy networks. To this end, we propose DopplerFi, a communication framework that enables a two-way communication channel between BLE and Wi-Fi by injecting artificial Doppler shifts, which can be decoded by sensing the patterns in the Gaussian frequency shift keying (GFSK) demodulator and Channel State Information (CSI). The artificial Doppler shifts can be compensated by the inherent frequency synchronization module and thus have a negligible impact on legacy communications. Our evaluation using commercial off-the-shelf (COTS) BLE chips and 802.11-compliant testbeds have demonstrated that DopplerFi can achieve throughput up to 6.5~Kbps at the cost of merely less than 0.8% throughput loss.

cs.NI↗
arXiv2018

Joint Neural Architecture Search and Quantization

Designing neural architectures is a fundamental step in deep learning applications. As a partner technique, model compression on neural networks has been widely investigated to gear the needs that the deep learning algorithms could be run with the limited computation resources on mobile devices. Currently, both the tasks of architecture design and model compression require expertise tricks and tedious trials. In this paper, we integrate these two tasks into one unified framework, which enables the joint architecture search with quantization (compression) policies for neural networks. This method is named JASQ. Here our goal is to automatically find a compact neural network model with high performance that is suitable for mobile devices. Technically, a multi-objective evolutionary search algorithm is introduced to search the models under the balance between model size and performance accuracy. In experiments, we find that our approach outperforms the methods that search only for architectures or only for quantization policies. 1) Specifically, given existing networks, our approach can provide them with learning-based quantization policies, and outperforms their 2 bits, 4 bits, 8 bits, and 16 bits counterparts. It can yield higher accuracies than the float models, for example, over 1.02% higher accuracy on MobileNet-v1. 2) What is more, under the balance between model size and performance accuracy, two models are obtained with joint search of architectures and quantization policies: a high-accuracy model and a small model, JASQNet and JASQNet-Small that achieves 2.97% error rate with 0.9 MB on CIFAR-10.

cs.CV↗
arXiv2018

A Minimization Method for The Double-Well Energy Functional

In this paper an iterative minimization method is proposed to approximate the minimizer to the double-well energy functional arising in the phase-field theory. The method is based on a quadratic functional posed over a nonempty closed convex set and is shown to be unconditionally energy stable. By the minimization approach, we also derive an variant of the first-order scheme for the Allen-Cahn equation, which has been constructed in the context of Invariant Energy Quadratization, and prove its unconditional energy stability.

math.NA↗
arXiv2018

In-the-wild Facial Expression Recognition in Extreme Poses

In the computer research area, facial expression recognition is a hot research problem. Recent years, the research has moved from the lab environment to in-the-wild circumstances. It is challenging, especially under extreme poses. But current expression detection systems are trying to avoid the pose effects and gain the general applicable ability. In this work, we solve the problem in the opposite approach. We consider the head poses and detect the expressions within special head poses. Our work includes two parts: detect the head pose and group it into one pre-defined head pose class; do facial expression recognize within each pose class. Our experiments show that the recognition results with pose class grouping are much better than that of direct recognition without considering poses. We combine the hand-crafted features, SIFT, LBP and geometric feature, with deep learning feature as the representation of the expressions. The handcrafted features are added into the deep learning framework along with the high level deep learning features. As a comparison, we implement SVM and random forest to as the prediction models. To train and test our methodology, we labeled the face dataset with 6 basic expressions.

cs.CV↗
arXiv2018

Boundary Hölder gradient estimates for the Monge-Ampère equation

We investigate global Hölder gradient estimates for solutions to the Monge-Ampère equation $$\mathrm{det}\;D^2 u=f\quad\mathrm{in}\;Ω,$$ where the right-hand side $f$ is bounded away from $0$ and $\infty$. We consider two main situations when a) the domain $Ω$ is uniformly convex and b) $Ω$ is flat.

math.AP↗
arXiv2018

ShieldScatter: Improving IoT Security with Backscatter Assistance

The lightweight protocols and low-power radio technologies open up many opportunities to facilitate Internet-of-Things (IoT) into our daily life, while their minimalist design also makes IoT devices vulnerable to many active attacks due to the lack of sophisticated security protocols. Recent advances advocate the use of an antenna array to extract fine-grained physical-layer signatures to mitigate these active attacks. However, it adds burdens in terms of energy consumption and hardware cost that IoT devices cannot afford. To overcome this predicament, we present ShieldScatter, a lightweight system that attaches battery-free backscatter tags to single-antenna devices to shield the system from active attacks. The key insight of ShieldScatter is to intentionally create multi-path propagation signatures with the careful deployment of backscatter tags. These signatures can be used to construct a sensitive profile to identify the location of the signals' arrival, and thus detect the threat. We prototype ShieldScatter with USRPs and ambient backscatter tags to evaluate our system in various environments. The experimental results show that even when the attacker is located only 15 cm away from the legitimate device, ShieldScatter with merely three backscatter tags can mitigate 97% of spoofing attack attempts while at the same time trigger false alarms on just 7% of legitimate traffic.

cs.CR↗
arXiv2018

Global $C^{1+α,\frac{1+α}{2}}$ regularity on the linearized parabolic Monge-Amp$\grave{e}$re equation

In this paper, we establish global $C^{1+α,\frac{1+α}{2}}$ estimates for solutions of the linearized parabolic Monge-Amp$\grave{e}$re equation $$\mathcal{L}_ϕu(x,t):=-u_t\,\mathrm{det}D^2ϕ(x)+\mathrm{tr}[Φ(x) D^2 u]=f(x,t)$$ under appropriate conditions on the domain, Monge-Amp$\grave{e}$re measures, boundary data and $f$, where $Φ:=\mathrm{det}(D^2ϕ)(D^2ϕ)^{-1}$ is the cofactor of the Hessian of $D^2ϕ$.

math.AP↗
arXiv2018

Interior $C^{1,α}$ regularity on the linearized Monge-Amp$\grave{e}$re equation with $\mathrm{VMO}$ type coefficients

In this paper, we establish interior $C^{1,α}$ estimates for solutions of the linearized Monge-Amp$\grave{e}$re equation $$\mathcal{L}_ϕu:=\mathrm{tr}[ΦD^2 u]=f,$$ where the density of the Monge-Amp$\grave{e}$re measure $g:=\mathrm{det}D^2ϕ$ satisfies a $\mathrm{VMO}$-type condition and $Φ:=(\mathrm{det}D^2ϕ)(D^2ϕ)^{-1}$ is the cofactor matrix of $D^2ϕ$.

math.AP↗
arXiv2018

Diffusion-controlled Alloying of Single-phase Multi-principal Covalent Transition Metal Carbides with Enhanced Damage tolerance and Exceptional Thermal Properties

Multicomponent alloying has displayed extraordinary potential for producing exceptional structural and functional materials. However, the synthesis of single-phase, multi-principal covalent compounds remains a challenge. Here we present a diffusion-controlled alloying strategy for the successful realization of covalent multi-principal transition metal carbides (MPTMCs) with a single face-centered cubic (FCC) phase. The increased interfacial diffusion promoted by the addition of a nonstoichiometric compound leads to rapid formation of the new single phase at much lower sintering temperature. Direct atomic-level observations via scanning transmission electron microscopy demonstrate that MPTMCs are composed of a single phase with a random distribution of all cations, which holds the key to the unique combinations of improved fracture toughness, superior Vickers hardness, and extremely lower thermal diffusivity achieved in MPTMCs. The present discovery provides a promising approach toward the design and synthesis of next-generation high-performance materials.

cond-mat.mtrl-sci↗
arXiv2018

Authenticating On-Body Backscatter by Exploiting Propagation Signatures

The vision of battery-free communication has made backscatter a compelling technology for on-body wearable and implantable devices. Recent advances have facilitated the communication between backscatter tags and on-body smart devices. These studies have focused on the communication dimension, while the security dimension remains vulnerable. It has been demonstrated that wireless connectivity can be exploited to send unauthorized commands or fake messages that result in device malfunctioning. The key challenge in defending these attacks stems from the minimalist design in backscatter. Thus, in this paper, we explore the feasibility of authenticating an on-body backscatter tag without modifying its signal or protocol. We present SecureScatter, a physical-layer solution that delegates the security of backscatter to an on-body smart device. To this end, we profile the on-body propagation paths of backscatter links, and construct highly sensitive propagation signatures to identify on-body backscatter links. We implement our design in a software radio and evaluate it with different backscatter tags that work at 2.4 GHz and 900 MHz. Results show that our system can identify on-body devices at 93.23% average true positive rate and 3.18% average false positive rate.

cs.HC↗
arXiv2018

Epidemic spreading on time-varying multiplex networks

Social interactions are stratified in multiple contexts and are subject to complex temporal dynamics. The systematic study of these two features of social systems has started only very recently mainly thanks to the development of multiplex and time-varying networks. However, these two advancements have progressed almost in parallel with very little overlap. Thus, the interplay between multiplexity and the temporal nature of connectivity patterns is poorly understood. Here, we aim to tackle this limitation by introducing a time-varying model of multiplex networks. We are interested in characterizing how these two properties affect contagion processes. To this end, we study SIS epidemic models unfolding at comparable time-scale respect to the evolution of the multiplex network. We study both analytically and numerically the epidemic threshold as a function of the overlap between, and the features of, each layer. We found that, the overlap between layers significantly reduces the epidemic threshold especially when the temporal activation patterns of overlapping nodes are positively correlated. Furthermore, when the average connectivity across layers is very different, the contagion dynamics are driven by the features of the more densely connected layer. Here, the epidemic threshold is equivalent to that of a single layered graph and the impact of the disease, in the layer driving the contagion, is independent of the overlap. However, this is not the case in the other layers where the spreading dynamics are sharply influenced by it. The results presented provide another step towards the characterization of the properties of real networks and their effects on contagion phenomena

physics.soc-ph↗
arXiv2018

Unsupervised Domain Adaptive Re-Identification: Theory and Practice

We study the problem of unsupervised domain adaptive re-identification (re-ID) which is an active topic in computer vision but lacks a theoretical foundation. We first extend existing unsupervised domain adaptive classification theories to re-ID tasks. Concretely, we introduce some assumptions on the extracted feature space and then derive several loss functions guided by these assumptions. To optimize them, a novel self-training scheme for unsupervised domain adaptive re-ID tasks is proposed. It iteratively makes guesses for unlabeled target data based on an encoder and trains the encoder based on the guessed labels. Extensive experiments on unsupervised domain adaptive person re-ID and vehicle re-ID tasks with comparisons to the state-of-the-arts confirm the effectiveness of the proposed theories and self-training framework. Our code is available at \url{https://github.com/LcDog/DomainAdaptiveReID}.

cs.CV↗