arXiv subjects

Zi Wang

Publications and source records attributed to Zi Wang.

At least 145 records · Page 8Linked to original sources

arXiv2021

Observation of Geometric Heat Pump Effect in Periodic Driven Thermal Diffusion

The concept of geometry works as an overarching framework underlying a wide range of transport phenomena. Particularly, the geometric phase effect in classical and quantum heat pump has been attracting much attention in microscopic systems. Here, we formulate theoretically the geometric heat pump effect in macroscopic driven diffusive systems. Upon modulation protocols, the nontrivial geometric curvature in the parameter space universally induces an additional pumped heat, beyond the constraint of hot-to-cold flowing. Furthermore, we set up a minimum experiment and indeed observe a non-vanishing directional heat flow across the driven system, despite keeping zero thermal bias between two time-dependent thermal reservoirs at every instant. We verify that in analogy to the geometric phase effect, the geometric pumped heat during each driving cycle is independent of driving periods in the adiabatic limit and coincides with theoretical predictions, thus validating its geometric origin. These results about geometric heat pump effect could have potential implications for designing and implementing nonreciprocal and topological thermal meta-devices under spatiotemporal modulations.

cond-mat.stat-mech↗

arXiv2021

Non-line-of-Sight Imaging via Neural Transient Fields

We present a neural modeling framework for Non-Line-of-Sight (NLOS) imaging. Previous solutions have sought to explicitly recover the 3D geometry (e.g., as point clouds) or voxel density (e.g., within a pre-defined volume) of the hidden scene. In contrast, inspired by the recent Neural Radiance Field (NeRF) approach, we use a multi-layer perceptron (MLP) to represent the neural transient field or NeTF. However, NeTF measures the transient over spherical wavefronts rather than the radiance along lines. We therefore formulate a spherical volume NeTF reconstruction pipeline, applicable to both confocal and non-confocal setups. Compared with NeRF, NeTF samples a much sparser set of viewpoints (scanning spots) and the sampling is highly uneven. We thus introduce a Monte Carlo technique to improve the robustness in the reconstruction. Comprehensive experiments on synthetic and real datasets demonstrate NeTF provides higher quality reconstruction and preserves fine details largely missing in the state-of-the-art.

eess.IV↗

arXiv2021

An Ear Canal Deformation Based User Authentication Using Ear Wearable Devices

Biometric-based authentication is gaining increasing attention for wearables and mobile applications. Meanwhile, the growing adoption of sensors in wearables also provides opportunities to capture novel wearable biometrics. In this work, we propose EarDynamic, an ear canal deformation-based user authentication using in-ear wearables. EarDynamic provides continuous and passive user authentication and is transparent to users. It leverages ear canal deformation that combines the unique static geometry and dynamic motions of the ear canal when the user is speaking for authentication. It utilizes an acoustic sensing approach to capture the ear canal deformation with the built-in microphone and speaker of the in-ear wearable. Specifically, it first emits well-designed inaudible beep signals and records the reflected signals from the ear canal. It then analyzes the reflected signals and extracts fine-grained acoustic features that correspond to the ear canal deformation for user authentication. Our extensive experimental evaluation shows that EarDynamic can achieve a recall of 97.38% and an F1 score of 96.84%. Results also show that our system works well under different noisy environments with various daily activities.

cs.CR↗

arXiv2021

Cycle Flux Ranking of Network Analysis in Quantum Thermal Device

Manipulating quantum thermal transport relies on uncovering the principle working cycles of quantum devices. Here, we apply the cycle flux ranking of network analysis to nonequilibrium thermal devices described by graphs of quantum state transitions. To excavate the principal mechanism out of complex transport behaviors, we decompose the quantum-transition network into cycles, calculate the cycle flux by algebraic graph theory, and pick out the dominant cycles with top-ranked fluxes, i.e., the cycle trajectories with highest probabilities. We demonstrate the cycle flux ranking in typical quantum device models, such as a thermal-drag spin-Seebeck pump, and a quantum thermal transistor as thermal switch or heat amplifier. The dominant cycle trajectories indeed elucidate the principal working mechanisms of those quantum devices. The cycle flux analysis provides an alternative perspective that naturally describes the working cycle corresponding to the main functionality of quantum thermal devices, which would further guide the device optimization with desired performance

quant-ph↗

arXiv2021

Interval Universal Approximation for Neural Networks

To verify safety and robustness of neural networks, researchers have successfully applied abstract interpretation, primarily using the interval abstract domain. In this paper, we study the theoretical power and limits of the interval domain for neural-network verification. First, we introduce the interval universal approximation (IUA) theorem. IUA shows that neural networks not only can approximate any continuous function $f$ (universal approximation) as we have known for decades, but we can find a neural network, using any well-behaved activation function, whose interval bounds are an arbitrarily close approximation of the set semantics of $f$ (the result of applying $f$ to a set of inputs). We call this notion of approximation interval approximation. Our theorem generalizes the recent result of Baader et al. (2020) from ReLUs to a rich class of activation functions that we call squashable functions. Additionally, the IUA theorem implies that we can always construct provably robust neural networks under $\ell_\infty$-norm using almost any practical activation function. Second, we study the computational complexity of constructing neural networks that are amenable to precise interval analysis. This is a crucial question, as our constructive proof of IUA is exponential in the size of the approximation domain. We boil this question down to the problem of approximating the range of a neural network with squashable activation functions. We show that the range approximation problem (RA) is a $Δ_2$-intermediate problem, which is strictly harder than $\mathsf{NP}$-complete problems, assuming $\mathsf{coNP}\not\subset \mathsf{NP}$. As a result, IUA is an inherently hard problem: No matter what abstract domain or computational tools we consider to achieve interval approximation, there is no efficient construction of such a universal approximator.

cs.LG↗

arXiv2021

Geometric Heat Pump: Controlling Thermal Transport with Time-dependent Modulations

The second law of thermodynamics dictates that heat simultaneously flows from the hot to cold bath on average. To go beyond this picture, a range of works in the past decade show that, other than the average dynamical heat flux determined by instantaneous thermal bias, a non-trivial flux contribution of intrinsic geometric origin is generally present in temporally driven systems. This additional heat flux provides a free lunch for the pumped heat and could even drive heat against the bias. We review here the emergence and development of this so called ``geometric heat pump'', originating from the topological geometric phase effect, and cover various quantum and classical transport systems with different internal dynamics. The generalization from the adiabatic to the non-adiabatic regime and the application of control theory are also discussed. Then, we briefly discuss the symmetry restriction on the heat pump effect, such as duality, supersymmetry and time-reversal symmetry. Finally, we examine open problems concerning the geometric heat pump process and elucidate their prospective significance in devising thermal machines with high performance.

cond-mat.stat-mech↗

arXiv2021

XCloud-pFISTA: A Medical Intelligence Cloud for Accelerated MRI

Machine learning and artificial intelligence have shown remarkable performance in accelerated magnetic resonance imaging (MRI). Cloud computing technologies have great advantages in building an easily accessible platform to deploy advanced algorithms. In this work, we develop an open-access, easy-to-use and high-performance medical intelligence cloud computing platform (XCloud-pFISTA) to reconstruct MRI images from undersampled k-space data. Two state-of-the-art approaches of the Projected Fast Iterative Soft-Thresholding Algorithm (pFISTA) family have been successfully implemented on the cloud. This work can be considered as a good example of cloud-based medical image reconstruction and may benefit the future development of integrated reconstruction and online diagnosis system.

eess.IV↗

arXiv2021

Zero-Shot Knowledge Distillation from a Decision-Based Black-Box Model

Knowledge distillation (KD) is a successful approach for deep neural network acceleration, with which a compact network (student) is trained by mimicking the softmax output of a pre-trained high-capacity network (teacher). In tradition, KD usually relies on access to the training samples and the parameters of the white-box teacher to acquire the transferred knowledge. However, these prerequisites are not always realistic due to storage costs or privacy issues in real-world applications. Here we propose the concept of decision-based black-box (DB3) knowledge distillation, with which the student is trained by distilling the knowledge from a black-box teacher (parameters are not accessible) that only returns classes rather than softmax outputs. We start with the scenario when the training set is accessible. We represent a sample's robustness against other classes by computing its distances to the teacher's decision boundaries and use it to construct the soft label for each training sample. After that, the student can be trained via standard KD. We then extend this approach to a more challenging scenario in which even accessing the training data is not feasible. We propose to generate pseudo samples distinguished by the teacher's decision boundaries to the largest extent and construct soft labels for them, which are used as the transfer set. We evaluate our approaches on various benchmark networks and datasets and experiment results demonstrate their effectiveness. Codes are available at: https://github.com/zwang84/zsdb3kd.

cs.LG↗

arXiv2021

Acoustic-based Object Detection for Pedestrian Using Smartphone

Walking while using a smartphone is becoming a major pedestrian safety concern as people may unknowingly bump into various obstacles that could lead to severe injuries. In this paper, we propose ObstacleWatch, an acoustic-based obstacle collision detection system to improve the safety of pedestrians who are engaged in smartphone usage while walking. ObstacleWatch leverages the advanced audio hardware of the smartphone to sense the surrounding obstacles and infers fine-grained information about the frontal obstacle for collision detection. In particular, our system emits well-designed inaudible beep signals from the smartphone built-in speaker and listens to the reflections with the stereo recording of the smartphone. By analyzing the reflected signals received at two microphones, ObstacleWatch is able to extract fine-grained information of the frontal obstacle including the distance, angle, and size for detecting the possible collisions and to alert users. Our experimental evaluation under two real-world environments with different types of phones and obstacles shows that ObstacleWatch achieves over 92% accuracy in predicting obstacle collisions with distance estimation errors at about 2 cm. Results also show that ObstacleWatch is robust to different sizes of objects and is compatible with different phone models with low energy consumption.

cs.HC↗

arXiv2021

Learning compositional models of robot skills for task and motion planning

The objective of this work is to augment the basic abilities of a robot by learning to use sensorimotor primitives to solve complex long-horizon manipulation problems. This requires flexible generative planning that can combine primitive abilities in novel combinations and thus generalize across a wide variety of problems. In order to plan with primitive actions, we must have models of the actions: under what circumstances will executing this primitive successfully achieve some particular effect in the world? We use, and develop novel improvements on, state-of-the-art methods for active learning and sampling. We use Gaussian process methods for learning the constraints on skill effectiveness from small numbers of expensive-to-collect training examples. Additionally, we develop efficient adaptive sampling methods for generating a comprehensive and diverse sequence of continuous candidate control parameter values (such as pouring waypoints for a cup) during planning. These values become end-effector goals for traditional motion planners that then solve for a full robot motion that performs the skill. By using learning and planning methods in conjunction, we take advantage of the strengths of each and plan for a wide variety of complex dynamic manipulation tasks. We demonstrate our approach in an integrated system, combining traditional robotics primitives with our newly learned models using an efficient robot task and motion planner. We evaluate our approach both in simulation and in the real world through measuring the quality of the selected primitive actions. Finally, we apply our integrated system to a variety of long-horizon simulated and real-world manipulation problems.

cs.RO↗

arXiv2021

Deep UV photolithography enhanced geometric homogeneity for low loss photonic crystal waveguides

Periodic or gradient subwavelength structures are basic configurations of photonic crystals and metamaterials. While the numerical simulations predict trivial loss, but the fabricated photonic crystal waveguides by E-beam lithography and standard dry etching shows unacceptable loss beyond 10 dB. Nanofabrication variation introduced geometric inhomogeneity is considered as the primary cause of the deleterious performance. The deep UV photolithography in CMOS foundry is a large-scale parallel processing, which can significantly suppress the fabrication-related inhomogeneous offsets and thus the optical linear loss. Here we demonstrate ultra-low loss photonic crystal waveguides with a 300 mm multi-project wafer run. For sub-millimetre long photonic crystal W1 waveguides, consistent performance of 2 dB total loss and 40 dB extinction ratio are observed across different dies.

physics.app-ph↗

arXiv2021

Data-Free Knowledge Distillation with Soft Targeted Transfer Set Synthesis

Knowledge distillation (KD) has proved to be an effective approach for deep neural network compression, which learns a compact network (student) by transferring the knowledge from a pre-trained, over-parameterized network (teacher). In traditional KD, the transferred knowledge is usually obtained by feeding training samples to the teacher network to obtain the class probabilities. However, the original training dataset is not always available due to storage costs or privacy issues. In this study, we propose a novel data-free KD approach by modeling the intermediate feature space of the teacher with a multivariate normal distribution and leveraging the soft targeted labels generated by the distribution to synthesize pseudo samples as the transfer set. Several student networks trained with these synthesized transfer sets present competitive performance compared to the networks trained with the original training set and other data-free KD approaches.

cs.LG↗

arXiv2021

Convolutional Neural Network Pruning with Structural Redundancy Reduction

Convolutional neural network (CNN) pruning has become one of the most successful network compression approaches in recent years. Existing works on network pruning usually focus on removing the least important filters in the network to achieve compact architectures. In this study, we claim that identifying structural redundancy plays a more essential role than finding unimportant filters, theoretically and empirically. We first statistically model the network pruning problem in a redundancy reduction perspective and find that pruning in the layer(s) with the most structural redundancy outperforms pruning the least important filters across all layers. Based on this finding, we then propose a network pruning approach that identifies structural redundancy of a CNN and prunes filters in the selected layer(s) with the most redundancy. Experiments on various benchmark network architectures and datasets show that our proposed approach significantly outperforms the previous state-of-the-art.

cs.CV↗

arXiv2021

Engineering the light coupling between metalens and photonic crystal resonators for robust on-chip microsystems

We designed an on-chip transformative optic system of a metalens-photonic crystal resonator metasystem on a foundry compatible silicon photonic platform. By adjusting the on-chip metalens' focusing length and mode dimension, the insertion loss between the metalens and the photonic crystal resonator and waveguide structures are minimized through mode-matching. The micro-system does not involve any single mode silicon nanowire waveguide, and thus mechanically robust without any oxide claddings. The proposed microsystem is ideal for miniaturized chemical and biosensors operating in air or solution environment.

physics.optics↗

arXiv2021

Exploring Adversarial Robustness of Deep Metric Learning

Deep Metric Learning (DML), a widely-used technique, involves learning a distance metric between pairs of samples. DML uses deep neural architectures to learn semantic embeddings of the input, where the distance between similar examples is small while dissimilar ones are far apart. Although the underlying neural networks produce good accuracy on naturally occurring samples, they are vulnerable to adversarially-perturbed samples that reduce performance. We take a first step towards training robust DML models and tackle the primary challenge of the metric losses being dependent on the samples in a mini-batch, unlike standard losses that only depend on the specific input-output pair. We analyze this dependence effect and contribute a robust optimization formulation. Using experiments on three commonly-used DML datasets, we demonstrate 5-76 fold increases in adversarial accuracy, and outperform an existing DML model that sought out to be robust.

cs.LG↗

arXiv2021

A Note on the Likelihood Ratio Test in High-Dimensional Exploratory Factor Analysis

The likelihood ratio test is widely used in exploratory factor analysis to assess the model fit and determine the number of latent factors. Despite its popularity and clear statistical rationale, researchers have found that when the dimension of the response data is large compared to the sample size, the classical chi-square approximation of the likelihood ratio test statistic often fails. Theoretically, it has been an open problem when such a phenomenon happens as the dimension of data increases; practically, the effect of high dimensionality is less examined in exploratory factor analysis, and there lacks a clear statistical guideline on the validity of the conventional chi-square approximation. To address this problem, we investigate the failure of the chi-square approximation of the likelihood ratio test in high-dimensional exploratory factor analysis, and derive the necessary and sufficient condition to ensure the validity of the chi-square approximation. The results yield simple quantitative guidelines to check in practice and would also provide useful statistical insights into the practice of exploratory factor analysis.

math.ST↗

arXiv2020

Review and Prospect: NMR Spectroscopy Denoising & Reconstruction with Low Rank Hankel Matrices and Tensors

Nuclear Magnetic Resonance (NMR) spectroscopy is an important analytical tool in chemistry, biology, and life science, but it suffers from relatively low sensitivity and long acquisition time. Thus, improving the apparent signal-to-noise ratio and accelerating data acquisition become indispensable. In this review, we summarize the recent progress on low rank Hankel matrix and tensor methods, that exploit the exponential property of free induction decay signals, to enable effective denoising and spectra reconstruction. We also outline future developments that are likely to make NMR spectroscopy a far more powerful technique.

physics.chem-ph↗

arXiv2020

Semantic Robustness of Models of Source Code

Deep neural networks are vulnerable to adversarial examples - small input perturbations that result in incorrect predictions. We study this problem for models of source code, where we want the network to be robust to source-code modifications that preserve code functionality. (1) We define a powerful adversary that can employ sequences of parametric, semantics-preserving program transformations; (2) we show how to perform adversarial training to learn models robust to such adversaries; (3) we conduct an evaluation on different languages and architectures, demonstrating significant quantitative gains in robustness.

cs.LG↗