arXiv subjects

Peng Chen

Publications and source records attributed to Peng Chen.

At least 361 records · Page 20Linked to original sources

arXiv2020

Projected Stein Variational Gradient Descent

The curse of dimensionality is a longstanding challenge in Bayesian inference in high dimensions. In this work, we propose a projected Stein variational gradient descent (pSVGD) method to overcome this challenge by exploiting the fundamental property of intrinsic low dimensionality of the data informed subspace stemming from ill-posedness of such problems. We adaptively construct the subspace using a gradient information matrix of the log-likelihood, and apply pSVGD to the much lower-dimensional coefficients of the parameter projection. The method is demonstrated to be more accurate and efficient than SVGD. It is also shown to be more scalable with respect to the number of parameters, samples, data points, and processor cores via experiments with parameters dimensions ranging from the hundreds to the tens of thousands.

cs.LG↗

arXiv2020

Audio-based automatic mating success prediction of giant pandas

Giant pandas, stereotyped as silent animals, make significantly more vocal sounds during breeding season, suggesting that sounds are essential for coordinating their reproduction and expression of mating preference. Previous biological studies have also proven that giant panda sounds are correlated with mating results and reproduction. This paper makes the first attempt to devise an automatic method for predicting mating success of giant pandas based on their vocal sounds. Given an audio sequence of mating giant pandas recorded during breeding encounters, we first crop out the segments with vocal sound of giant pandas, and normalize its magnitude, and length. We then extract acoustic features from the audio segment and feed the features into a deep neural network, which classifies the mating into success or failure. The proposed deep neural network employs convolution layers followed by bidirection gated recurrent units to extract vocal features, and applies attention mechanism to force the network to focus on most relevant features. Evaluation experiments on a data set collected during the past nine years obtain promising results, proving the potential of audio-based automatic mating success prediction methods in assisting giant panda reproduction.

cs.SD↗

arXiv2020

Massive suppression of proximity pairing in topological (Bi$_{1-x}$Sb$_{x})_2$Te$_3$ films on niobium

Interfacing bulk conducting topological Bi$_2$Se$_3$ films with s-wave superconductors initiates strong superconducting order in the nontrivial surface states. However, bulk insulating topological (Bi$_{1-x}$Sb$_{x})_2$Te$_3$ films on bulk Nb instead exhibit a giant attenuation of surface superconductivity, even for films only two-layers thick. This massive suppression of proximity pairing is evidenced by ultrahigh-resolution band mappings and by contrasting quantified superconducting gaps with those of heavily n-doped topological Bi$_2$Se$_3$/Nb. The results underscore the limitations of using superconducting proximity effects to realize topological superconductivity in nearly intrinsic systems.

cond-mat.supr-con↗

arXiv2020

Ferroelectric 180 degree walls are mechanically softer than the domains they separate

Domain walls are functionally different from the domains they separate, but little is known about their mechanical properties. Using scanning probe microscopy, we have measured the mechanical response of ferroelectric 180o domain walls and observed that, despite separating domains that are mechanically identical (non-ferroelastic), the walls are mechanically distinct -- softer -- compared to the domains. This effect has been observed in different ferroelectric materials (LiNbO3, BaTiO3, PbTiO3) and with different morphologies (from single crystals to thin films) so it appears to be universal. We propose a theoretical framework that explains the domain wall softening and justifies that the effect should be common to all ferroelectrics.

cond-mat.mtrl-sci↗

arXiv2020

Sharp endpoint $L^p$ estimates for Schrödinger groups

Let $L$ be a non-negative self-adjoint operator acting on $L^2(X)$ where $X$ is a space of homogeneous type with a dimension $n$. Suppose that the heat operator $e^{-tL}$ satisfies the generalized Gaussian $(p_0, p'_0)$-estimates of order $m$ for some $1\leq p_0 < 2$. In this paper we prove {\it sharp} endpoint $L^p$-Sobolev bound for the Schrödinger group $e^{itL}$, that is for every $p\in (p_0, p'_0)$ there exists a constant $C=C(n,p)>0$ independent of $t$ such that \begin{eqnarray*} \left\| (I+L)^{-{s}}e^{itL} f\right\|_{p} \leq C(1+|t|)^{s}\|f\|_{p}, \ \ \ t\in{\mathbb R}, \ \ \ s\geq n\big|{1\over 2}-{1\over p}\big|. \end{eqnarray*} As a consequence, the above estimate holds for all $1<p<\infty$ when the heat kernel of $L$ satisfies a Gaussian upper bound. This extends classical results due to Feffermann and Stein, and Miyachi for the Laplacian on the Euclidean spaces ${\mathbb R}^n$. We also give an application to obtain an endpoint estimate for $L^p$-boundedness of the Riesz means of the solutions of the Schrödinger equations.

math.CA↗

arXiv2020

Deep Neural Network-Based Quantized Signal Reconstruction for DOA Estimation

For a massive multiple-input-multiple-output (MIMO) system using intelligent reflecting surface (IRS) equipped with radio frequency (RF) chains, the multi-channel RF chains are expensive compared to passive IRS, especially, when the high-resolution and high-speed analog to digital converters (ADC) are used in each RF channel. In this letter, a direction of angle (DOA) estimation problem is investigated with low-cost ADC in IRS, and we propose a deep neural network (DNN) as a recovery method for the low-resolution sampled signal. Different from the existing denoising convolutional neural network (DnCNN) for Gaussian noise, the proposed DNN with fully connected (FC) layers estimates the quantization noise caused by the ADC. Then, the denoised signal is subjected to the DOA estimation, and the recovery performance for the quantized signal is evaluated by DOA estimation. Simulation results show that under the same training conditions, the better reconstruction performance is achieved by the proposed network than state-of-the-art methods. The performance of the DOA estimation using 1-bit ADC is improved to exceed that using 2-bit ADC.

eess.SP↗

arXiv2020

Stein variational reduced basis Bayesian inversion

We propose and analyze a Stein variational reduced basis method (SVRB) to solve large-scale PDE-constrained Bayesian inverse problems. To address the computational challenge of drawing numerous samples requiring expensive PDE solves from the posterior distribution, we integrate an adaptive and goal-oriented model reduction technique with an optimization-based Stein variational gradient descent method (SVGD). The samples are drawn from the prior distribution and iteratively pushed to the posterior by a sequence of transport maps, which are constructed by SVGD, requiring the evaluation of the potential---the negative log of the likelihood function---and its gradient with respect to the random parameters, which depend on the solution of the PDE. To reduce the computational cost, we develop an adaptive and goal-oriented model reduction technique based on reduced basis approximations for the evaluation of the potential and its gradient. We present a detailed analysis for the reduced basis approximation errors of the potential and its gradient, the induced errors of the posterior distribution measured by Kullback--Leibler divergence, as well as the errors of the samples. To demonstrate the computational accuracy and efficiency of SVRB, we report results of numerical experiments on a Bayesian inverse problem governed by a diffusion PDE with random parameters with both uniform and Gaussian prior distributions. Over 100X speedups can be achieved while the accuracy of the approximation of the potential and its gradient is preserved.

math.NA↗

arXiv2020

An Efficient Secure Dynamic Skyline Query Model

It is now cost-effective to outsource large dataset and perform query over the cloud. However, in this scenario, there exist serious security and privacy issues that sensitive information contained in the dataset can be leaked. The most effective way to address that is to encrypt the data before outsourcing. Nevertheless, it remains a grand challenge to process queries in ciphertext efficiently. In this work, we shall focus on solving one representative query task, namely dynamic skyline query, in a secure manner over the cloud. However, it is difficult to be performed on encrypted data as its dynamic domination criteria require both subtraction and comparison, which cannot be directly supported by a single encryption scheme efficiently. To this end, we present a novel framework called SCALE. It works by transforming traditional dynamic skyline domination into pure comparisons. The whole process can be completed in single-round interaction between user and the cloud. We theoretically prove that the outsourced database, query requests, and returned results are all kept secret under our model. Moreover, we also present an efficient strategy for dynamic insertion and deletion of stored records. Empirical study over a series of datasets demonstrates that our framework improves the efficiency of query processing by nearly three orders of magnitude compared to the state-of-the-art.

cs.DB↗

arXiv2020

Liberation of slave modes inside domain walls in multiferroic Cu-Cl boracite

Domain walls (DWs), the two-dimensional boundaries between symmetry equivalent ferroic domains, are actively investigated due to their promise for novel logic and memory devices. Moreover, they can be easily created, erased and reshaped at a low energy cost due to their high mobility and large electrical conductivity. Most work so far has been focused on DWs in proper ferroelectrics, where the primary order parameter, ferroelectric polarization, interpolates between the values in the domains by either reducing to zero (in Ising-type DW) or rotating (Bloch type DW). Here we present a new member of DW family with a complex inner texture of slave order parameters inside the wall where the primary order parameter reduces to zero. Our first-principles-derived model predicts the existence of monopolar and toroidal polarization patterns. The results enable large-scale phase field simulations of complex domain patterns in boracites and could inspire novel devices based on domain walls in improper ferroelectrics.

cond-mat.mtrl-sci↗

arXiv2020

Projected Stein Variational Newton: A Fast and Scalable Bayesian Inference Method in High Dimensions

We propose a fast and scalable variational method for Bayesian inference in high-dimensional parameter space, which we call projected Stein variational Newton (pSVN) method. We exploit the intrinsic low-dimensional geometric structure of the posterior distribution in the high-dimensional parameter space via its Hessian (of the log posterior) operator and perform a parallel update of the parameter samples projected into a low-dimensional subspace by an SVN method. The subspace is adaptively constructed using the eigenvectors of the averaged Hessian at the current samples. We demonstrate fast convergence of the proposed method and its scalability with respect to the number of parameters, samples, and processor cores.

math.OC↗

arXiv2020

Training-free Monocular 3D Event Detection System for Traffic Surveillance

We focus on the problem of detecting traffic events in a surveillance scenario, including the detection of both vehicle actions and traffic collisions. Existing event detection systems are mostly learning-based and have achieved convincing performance when a large amount of training data is available. However, in real-world scenarios, collecting sufficient labeled training data is expensive and sometimes impossible (e.g. for traffic collision detection). Moreover, the conventional 2D representation of surveillance views is easily affected by occlusions and different camera views in nature. To deal with the aforementioned problems, in this paper, we propose a training-free monocular 3D event detection system for traffic surveillance. Our system firstly projects the vehicles into the 3D Euclidean space and estimates their kinematic states. Then we develop multiple simple yet effective ways to identify the events based on the kinematic patterns, which need no further training. Consequently, our system is robust to the occlusions and the viewpoint changes. Exclusive experiments report the superior result of our method on large-scale real-world surveillance datasets, which validates the effectiveness of our proposed system.

cs.CV↗

arXiv2020

Unexpected Giant Microwave Conductivity in a Nominally Silent BiFeO3 Domain Wall

Nanoelectronic devices based on ferroelectric domain walls (DWs), such as memories, transistors, and rectifiers, have been demonstrated in recent years. Practical high-speed electronics, on the other hand, usually demand operation frequencies in the giga-Hertz (GHz) regime, where the effect of dipolar oscillation is important. In this work, an unexpected giant GHz conductivity on the order of 103 S/m is observed in certain BiFeO3 DWs, which is about 100,000 times greater than the carrier-induced dc conductivity of the same walls. Surprisingly, the nominal configuration of the DWs precludes the ac conduction under an excitation electric field perpendicular to the surface. Theoretical analysis shows that the inclined DWs are stressed asymmetrically near the film surface, whereas the vertical walls in a control sample are not. The resultant imbalanced polarization profile can then couple to the out-of-plane microwave fields and induce power dissipation, which is confirmed by the phase-field modeling. Since the contributions from mobile-carrier conduction and bound-charge oscillation to the ac conductivity are equivalent in a microwave circuit, the research on local structural dynamics may open a new avenue to implement DW nano-devices for RF applications.

cond-mat.mtrl-sci↗

arXiv2019

One-Shot Imitation Filming of Human Motion Videos

Imitation learning has been applied to mimic the operation of a human cameraman in several autonomous cinematography systems. To imitate different filming styles, existing methods train multiple models, where each model handles a particular style and requires a significant number of training samples. As a result, existing methods can hardly generalize to unseen styles. In this paper, we propose a framework, which can imitate a filming style by "seeing" only a single demonstration video of the same style, i.e., one-shot imitation filming. This is done by two key enabling techniques: 1) feature extraction of the filming style from the demo video, and 2) filming style transfer from the demo video to the new situation. We implement the approach with deep neural network and deploy it to a 6 degrees of freedom (DOF) real drone cinematography system by first predicting the future camera motions, and then converting them to the drone's control commands via an odometer. Our experimental results on extensive datasets and showcases exhibit significant improvements in our approach over conventional baselines and our approach can successfully mimic the footage with an unseen style.

cs.CV↗

arXiv2019

Symmetry-breaking and spin-blockage effects on carrier dynamics in single-layer tungsten diselenide

Understanding carrier creation and evolution in materials initiated by pulsed optical excitation is central to developing ultrafast optoelectronics. We demonstrate herein that the dynamic response of a system can be drastically modified when its physical dimension is reduced to the atomic scale, the ultimate limit of device miniaturization. A comparative study of single-layer (SL) tungsten diselenide(WSe2) relative to bulk WSe2 shows substantial differences in the transient response as measured by time- and angle-resolved photoemission spectroscopy (TRARPES). The conduction-band minimum in bulk WSe2, populated by optical pumping, decays promptly. The corresponding decay for SL WSe2 is much slower and exhibits two time constants. The results indicate the presence of two distinct decay channels in the SL that are correlated with the breaking of space inversion symmetry in the two-dimensional limit. This symmetry breaking lifts the spin degeneracy of the bands, which in turn causes the blockage of decay for one spin channel. The stark contrast between the single layer and the bulk illustrates the basic carrier scattering processes operating at different timescales that can be substantially modified by dimensional and symmetry-reduction effects.

cond-mat.mtrl-sci↗

arXiv2019

Tailoring Hybrid Anomalous Hall Response in Engineered Magnetic Topological Insulator Heterostructures

Engineering the anomalous Hall effect (AHE) in the emerging magnetic topological insulators (MTIs) has great potentials for quantum information processing and spintronics applications. In this letter, we synthesize the epitaxial Bi2Te3/MnTe magnetic heterostructures and observe pronounced AHE signals from both layers combined together. The evolution of the resulting hybrid AHE intensity with the top Bi2Te3 layer thickness manifests the presence of an intrinsic ferromagnetic phase induced by the topological surface states at the heterolayer-interface. More importantly, by doping the Bi2Te3 layer with Sb, we are able to manipulate the sign of the Berry phase-associated AHE component. Our results demonstrate the un-paralleled advantages of MTI heterostructures over magnetically doped TI counterparts, in which the tunability of the AHE response can be greatly enhanced. This in turn unveils a new avenue for MTI heterostructure-based multifunctional applications.

cond-mat.mes-hall↗

arXiv2019

DLA: Dense-Layer-Analysis for Adversarial Example Detection

In recent years Deep Neural Networks (DNNs) have achieved remarkable results and even showed super-human capabilities in a broad range of domains. This led people to trust in DNNs' classifications and resulting actions even in security-sensitive environments like autonomous driving. Despite their impressive achievements, DNNs are known to be vulnerable to adversarial examples. Such inputs contain small perturbations to intentionally fool the attacked model. In this paper, we present a novel end-to-end framework to detect such attacks during classification without influencing the target model's performance. Inspired by recent research in neuron-coverage guided testing we show that dense layers of DNNs carry security-sensitive information. With a secondary DNN we analyze the activation patterns of the dense layers during classification runtime, which enables effective and real-time detection of adversarial examples. Our prototype implementation successfully detects adversarial examples in image, natural language, and audio processing. Thereby, we cover a variety of target DNNs, including Long Short Term Memory (LSTM) architectures. In addition, to effectively defend against state-of-the-art attacks, our approach generalizes between different sets of adversarial examples. Thus, our method most likely enables us to detect even future, yet unknown attacks. Finally, during white-box adaptive attacks, we show our method cannot be easily bypassed.

cs.CR↗

arXiv2019

Matryoshka: Fuzzing Deeply Nested Branches

Greybox fuzzing has made impressive progress in recent years, evolving from heuristics-based random mutation to approaches for solving individual path constraints. However, they have difficulty solving path constraints that involve deeply nested conditional statements, which are common in image and video decoders, network packet analyzers, and checksum tools. We propose an approach for addressing this problem. First, we identify all the control flow-dependent conditional statements of the target conditional statement. Next, we select the data flow-dependent conditional statements. Finally, we use three strategies to find an input that satisfies all conditional statements simultaneously. We implemented this approach in a tool called Matryoshka and compared its effectiveness on 13 open source programs against other state-of-the-art fuzzers. Matryoshka found significantly more unique crashes than AFL, QSYM, and Angora. We manually classified those crashes into 41 unique new bugs, and obtained 12 CVEs. Our evaluation also uncovered the key technique contributing to Matryoshka's impressive performance: it collects only the nesting constraints that may cause the target conditional statements unreachable, which greatly simplifies the constraints that it has to solve.

cs.CR↗

arXiv2019

Futures-Based Resource Trading and Fair Pricing in Real-Time IoT Networks

Technology advancements on sensing, communications, and computing directly accelerate the recent development of Internet of Things (IoT), leading to rich and diverse applications in industry and business processes. For enabling IoT applications, smart devices and networks rely on the capabilities for information collection, processing and tight collaboration with their neighbors. However, the rapid proliferation of IoT devices and real-time IoT applications result in dramatically increased demand for communication and computing resources, posing a significant challenge for the underlying resource constrained wireless networks. Due to their resource constraint, it is extremely difficult for such devices and networks to meet stringent requirements of many real-time IoT applications. To overcome these challenges, resource sharing among IoT devices and networks has been studied as an effective solution by exploiting unused resources.

cs.NI↗