arXiv subjects

Tao Zhou

Publications and source records attributed to Tao Zhou.

At least 235 records · Page 13Linked to original sources

arXiv2022

Adaptive deep density approximation for fractional Fokker-Planck equations

In this work, we propose adaptive deep learning approaches based on normalizing flows for solving fractional Fokker-Planck equations (FPEs). The solution of a FPE is a probability density function (PDF). Traditional mesh-based methods are ineffective because of the unbounded computation domain, a large number of dimensions and the nonlocal fractional operator. To this end, we represent the solution with an explicit PDF model induced by a flow-based deep generative model, simplified KRnet, which constructs a transport map from a simple distribution to the target distribution. We consider two methods to approximate the fractional Laplacian. One method is the Monte Carlo approximation. The other method is to construct an auxiliary model with Gaussian radial basis functions (GRBFs) to approximate the solution such that we may take advantage of the fact that the fractional Laplacian of a Gaussian is known analytically. Based on these two different ways for the approximation of the fractional Laplacian, we propose two models, MCNF and GRBFNF, to approximate stationary FPEs and MCTNF to approximate time-dependent FPEs. To further improve the accuracy, we refine the training set and the approximate solution alternately. A variety of numerical examples is presented to demonstrate the effectiveness of our adaptive deep density approaches.

cs.LG↗

arXiv2022

Scanning X-ray Diffraction Microscopy for Diamond Quantum Sensing

Understanding nano- and micro-scale crystal strain in CVD diamond is crucial to the advancement of diamond quantum technologies. In particular, the presence of such strain and its characterization present a challenge to diamond-based quantum sensing and information applications -- as well as for future dark matter detectors where directional information of incoming particles is encoded in crystal strain. Here, we exploit nanofocused scanning X-ray diffraction microscopy to quantitatively measure crystal deformation from defects in diamond with high spatial and strain resolution. Combining information from multiple Bragg angles allows stereoscopic three-dimensional modeling of strain feature geometry; the diffraction results are validated via comparison to optical measurements of the strain tensor based on spin-state-dependent spectroscopy of ensembles of nitrogen vacancy (NV) centers in the diamond. Our results demonstrate both strain and spatial resolution sufficient for directional detection of dark matter via X-ray measurement of crystal strain, and provide a promising tool for diamond growth analysis and improvement of defect-based sensing.

physics.app-ph↗

arXiv2022

Dynamic Tilting of Ferroelectric Domain Walls via Optically Induced Electronic Screening

Optical excitation perturbs the balance of phenomena selecting the tilt orientation of domain walls within ferroelectric thin films. The high carrier density induced in a low-strain BaTiO3 thin film by an above-bandgap ultrafast optical pulse changes the tilt angle that 90° a/c domain walls form with respect to the substrate-film interface. The dynamics of the changes are apparent in time-resolved synchrotron x-ray scattering studies of the domain diffuse scattering. Tilting occurs at 298 K, a temperature at which the a/b and a/c domain phases coexist but is absent at 343 K in the better ordered single-phase a/c regime. Phase coexistence at 298 K leads to increased domain-wall charge density, and thus a larger screening effect than in the single-phase regime. The screening mechanism points to new directions for the manipulation of nanoscale ferroelectricity.

cond-mat.mtrl-sci↗

arXiv2022

Prompt-driven efficient Open-set Semi-supervised Learning

Open-set semi-supervised learning (OSSL) has attracted growing interest, which investigates a more practical scenario where out-of-distribution (OOD) samples are only contained in unlabeled data. Existing OSSL methods like OpenMatch learn an OOD detector to identify outliers, which often update all modal parameters (i.e., full fine-tuning) to propagate class information from labeled data to unlabeled ones. Currently, prompt learning has been developed to bridge gaps between pre-training and fine-tuning, which shows higher computational efficiency in several downstream tasks. In this paper, we propose a prompt-driven efficient OSSL framework, called OpenPrompt, which can propagate class information from labeled to unlabeled data with only a small number of trainable parameters. We propose a prompt-driven joint space learning mechanism to detect OOD data by maximizing the distribution gap between ID and OOD samples in unlabeled data, thereby our method enables the outliers to be detected in a new way. The experimental results on three public datasets show that OpenPrompt outperforms state-of-the-art methods with less than 1% of trainable parameters. More importantly, OpenPrompt achieves a 4% improvement in terms of AUROC on outlier detection over a fully supervised model on CIFAR10.

cs.CV↗

arXiv2022

Myopia prediction for adolescents via time-aware deep learning

Background: Quantitative prediction of the adolescents' spherical equivalent based on their variable-length historical vision records. Methods: From October 2019 to March 2022, we examined binocular uncorrected visual acuity, axial length, corneal curvature, and axial of 75,172 eyes from 37,586 adolescents aged 6-20 years in Chengdu, China. 80\% samples consist of the training set and the remaining 20\% form the testing set. Time-Aware Long Short-Term Memory was used to quantitatively predict the adolescents' spherical equivalent within two and a half years. Result: The mean absolute prediction error on the testing set was 0.273-0.257 for spherical equivalent, ranging from 0.189-0.160 to 0.596-0.473 if we consider different lengths of historical records and different prediction durations. Conclusions: Time-Aware Long Short-Term Memory was applied to captured the temporal features in irregularly sampled time series, which is more in line with the characteristics of real data and thus has higher applicability, and helps to identify the progression of myopia earlier. The overall error 0.273 is much smaller than the criterion for clinically acceptable prediction, say 0.75.

cs.LG↗

arXiv2022

Deep learning at the edge enables real-time streaming ptychographic imaging

Coherent microscopy techniques provide an unparalleled multi-scale view of materials across scientific and technological fields, from structural materials to quantum devices, from integrated circuits to biological cells. Driven by the construction of brighter sources and high-rate detectors, coherent X-ray microscopy methods like ptychography are poised to revolutionize nanoscale materials characterization. However, associated significant increases in data and compute needs mean that conventional approaches no longer suffice for recovering sample images in real-time from high-speed coherent imaging experiments. Here, we demonstrate a workflow that leverages artificial intelligence at the edge and high-performance computing to enable real-time inversion on X-ray ptychography data streamed directly from a detector at up to 2 kHz. The proposed AI-enabled workflow eliminates the sampling constraints imposed by traditional ptychography, allowing low dose imaging using orders of magnitude less data than required by traditional methods.

cs.LG↗

arXiv2022

Discriminating abilities of threshold-free evaluation metrics in link prediction

Link prediction is a paradigmatic and challenging problem in network science, which attempts to uncover missing links or predict future links, based on known topology. A fundamental but still unsolved issue is how to choose proper metrics to fairly evaluate prediction algorithms. The area under the receiver operating characteristic curve (AUC) and the balanced precision (BP) are the two most popular metrics in early studies, while their effectiveness is recently under debate. At the same time, the area under the precision-recall curve (AUPR) becomes increasingly popular, especially in biological studies. Based on a toy model with tunable noise and predictability, we propose a method to measure the discriminating abilities of any given metric. We apply this method to the above three threshold-free metrics, showing that AUC and AUPR are remarkably more discriminating than BP, and AUC is slightly more discriminating than AUPR. The result suggests that it is better to simultaneously use AUC and AUPR in evaluating link prediction algorithms, at the same time, it warns us that the evaluation based only on BP may be unauthentic. This article provides a starting point towards a comprehensive picture about effectiveness of evaluation metrics for link prediction and other classification problems.

physics.data-an↗

arXiv2022

A predictor-corrector deep learning algorithm for high dimensional stochastic partial differential equations

In this paper, we present a deep learning-based numerical method for approximating high dimensional stochastic partial differential equations (SPDEs). At each time step, our method relies on a predictor-corrector procedure. More precisely, we decompose the original SPDE into a degenerate SPDE and a deterministic PDE. Then in the prediction step, we solve the degenerate SPDE with the Euler scheme, while in the correction step we solve the second-order deterministic PDE by deep neural networks via its equivalent backward stochastic differential equation (BSDE). Under standard assumptions, error estimates and the rate of convergence of the proposed algorithm are presented. The efficiency and accuracy of the proposed algorithm are illustrated by numerical examples.

math.NA↗

arXiv2022

Equivalence between Time Series Predictability and Bayes Error Rate

Predictability is an emerging metric that quantifies the highest possible prediction accuracy for a given time series, being widely utilized in assessing known prediction algorithms and characterizing intrinsic regularities in human behaviors. Lately, increasing criticisms aim at the inaccuracy of the estimated predictability, caused by the original entropy-based method. In this brief report, we strictly prove that the time series predictability is equivalent to a seemingly unrelated metric called Bayes error rate that explores the lowest error rate unavoidable in classification. This proof bridges two independently developed fields, and thus each can immediately benefit from the other. For example, based on three theoretical models with known and controllable upper bounds of prediction accuracy, we show that the estimation based on Bayes error rate can largely solve the inaccuracy problem of predictability.

cs.IT↗

arXiv2022

A lightweight, user-configurable detector ASIC digital architecture with on-chip data compression for MHz X-ray coherent diffraction imaging

Today, most X-ray pixel detectors used at light sources transmit raw pixel data off the detector ASIC. With the availability of more advanced ASIC technology nodes for scientific application, more digital functionality from the computing domains (e.g., compression) can be integrated directly into a detector ASIC to increase data velocity. In this paper, we describe a lightweight, user-configurable detector ASIC digital architecture with on-chip compression which can be implemented in \SI{130}{\nm} technologies in a reasonable area on the ASIC periphery. In addition, we present a design to efficiently handle the variable data from the stream of parallel compressors. The architecture includes user-selectable lossy and lossless compression blocks. The impact of lossy compression algorithms is evaluated on simulated and experimental X-ray ptychography datasets. This architecture is a practical approach to increase pixel detector frame rates towards the continuous \SI{1}{\MHz} regime for not only coherent imaging techniques such as ptychography, but also for other diffraction techniques at X-ray light sources.

physics.ins-det↗

arXiv2022

Camouflaged Object Detection via Context-aware Cross-level Fusion

Camouflaged object detection (COD) aims to identify the objects that conceal themselves in natural scenes. Accurate COD suffers from a number of challenges associated with low boundary contrast and the large variation of object appearances, e.g., object size and shape. To address these challenges, we propose a novel Context-aware Cross-level Fusion Network (C2F-Net), which fuses context-aware cross-level features for accurately identifying camouflaged objects. Specifically, we compute informative attention coefficients from multi-level features with our Attention-induced Cross-level Fusion Module (ACFM), which further integrates the features under the guidance of attention coefficients. We then propose a Dual-branch Global Context Module (DGCM) to refine the fused features for informative feature representations by exploiting rich global context information. Multiple ACFMs and DGCMs are integrated in a cascaded manner for generating a coarse prediction from high-level features. The coarse prediction acts as an attention map to refine the low-level features before passing them to our Camouflage Inference Module (CIM) to generate the final prediction. We perform extensive experiments on three widely used benchmark datasets and compare C2F-Net with state-of-the-art (SOTA) models. The results show that C2F-Net is an effective COD model and outperforms SOTA models remarkably. Further, an evaluation on polyp segmentation datasets demonstrates the promising potentials of our C2F-Net in COD downstream applications. Our code is publicly available at: https://github.com/Ben57882/C2FNet-TSCVT.

cs.CV↗

arXiv2022

Personalized recommendation system based on social relationships and historical behaviors

Previous studies show that recommendation algorithms based on historical behaviors of users can provide satisfactory recommendation performance. Many of these algorithms pay attention to the interest of users, while ignore the influence of social relationships on user behaviors. Social relationships not only carry intrinsic information of similar consumption tastes or behaviors, but also imply the influence of individual to its neighbors. In this paper, we assume that social relationships and historical behaviors of users are related to the same factors. Based on this assumption, we propose an algorithm to focus on social relationships useful for recommendation systems through mutual constraints from both types of information. We test the performance of our algorithm on four types of users, including all users, active users, inactive users and cold-start users. Results show that the proposed algorithm outperforms benchmarks in four types of scenarios subject to recommendation accuracy and diversity metrics. We further design a randomization model to explore the contribution of social relationships to recommendation performance, and the result shows that the contribution of social relationships in the proposed algorithm depends on the coupling strength of social relationships and historical behaviors.

cs.SI↗

arXiv2022

RGB-D Salient Object Detection: A Survey

Salient object detection (SOD), which simulates the human visual perception system to locate the most attractive object(s) in a scene, has been widely applied to various computer vision tasks. Now, with the advent of depth sensors, depth maps with affluent spatial information that can be beneficial in boosting the performance of SOD, can easily be captured. Although various RGB-D based SOD models with promising performance have been proposed over the past several years, an in-depth understanding of these models and challenges in this topic remains lacking. In this paper, we provide a comprehensive survey of RGB-D based SOD models from various perspectives, and review related benchmark datasets in detail. Further, considering that the light field can also provide depth maps, we review SOD models and popular benchmark datasets from this domain as well. Moreover, to investigate the SOD ability of existing models, we carry out a comprehensive evaluation, as well as attribute-based evaluation of several representative RGB-D based SOD models. Finally, we discuss several challenges and open directions of RGB-D based SOD for future research. All collected models, benchmark datasets, source code links, datasets constructed for attribute-based evaluation, and codes for evaluation will be made publicly available at https://github.com/taozh2017/RGBDSODsurvey

cs.CV↗

arXiv2022

The Gender Pay Gap in China: Insights from a Discrimination Perspective

Equal pay is an essential component of gender equality, one of the Sustainable Development Goals of the United Nations. Using resume data of over ten million Chinese online job seekers in 2015, we study the current gender pay gap in China. The results show that on average women only earned 71.57\% of what men earned in China. The gender pay gap exists across all age groups and educational levels. Contrary to the commonly held view that developments in education, economy, and a more open culture would reduce the gender pay gap, the fusion analysis of resume data and socio-economic data presents that they have not helped reach the gender pay equality in China. China seems to be stuck in a place where traditional methods cannot make further progress. Our analysis further shows that 81.47\% of the variance in the gender pay gap can be potentially attributed to discrimination. In particular, compared with the unmarried, both the gender pay gap itself and proportion potentially attributed to discrimination of the married are larger, indicating that married women suffer greater inequality and more discrimination than unmarried ones. Taken together, we suggest that more research attention should be paid to the effect of discrimination in understanding gender pay gap based on the family constraint theory. We also suggest the Chinese government to increase investment in family-supportive policies and grants in addition to female education.

physics.soc-ph↗

arXiv2022

An augmented Lagrangian deep learning method for variational problems with essential boundary conditions

This paper is concerned with a novel deep learning method for variational problems with essential boundary conditions. To this end, we first reformulate the original problem into a minimax problem corresponding to a feasible augmented Lagrangian, which can be solved by the augmented Lagrangian method in an infinite dimensional setting. Based on this, by expressing the primal and dual variables with two individual deep neural network functions, we present an augmented Lagrangian deep learning method for which the parameters are trained by the stochastic optimization method together with a projection technique. Compared to the traditional penalty method, the new method admits two main advantages: i) the choice of the penalty parameter is flexible and robust, and ii) the numerical solution is more accurate in the same magnitude of computational cost. As typical applications, we apply the new approach to solve elliptic problems and (nonlinear) eigenvalue problems with essential boundary conditions, and numerical experiments are presented to show the effectiveness of the new method.

math.NA↗

arXiv2022

Discrete energy analysis of the third-order variable-step BDF time-stepping for diffusion equations

This is one of our series works on discrete energy analysis of the variable-step BDF schemes. In this part, we present stability and convergence analysis of the third-order BDF (BDF3) schemes with variable steps for linear diffusion equations, see e.g. [SIAM J. Numer. Anal., 58:2294-2314] and [Math. Comp., 90: 1207-1226] for our previous works on the BDF2 scheme. To this aim, we first build up a discrete gradient structure of the variable-step BDF3 formula under the condition that the adjacent step ratios are less than 1.4877, by which we can establish a discrete energy dissipation law. Mesh-robust stability and convergence analysis in the $L^2$ norm are then obtained. Here the mesh robustness means that the solution errors are well controlled by the maximum time-step size but independent of the adjacent time-step ratios. We also present numerical tests to support our theoretical results.

math.NA↗

arXiv2022

Threshold solutions for the focusing generalized Hartree equations

We study the global behavior of solutions to the focusing generalized Hartree equation with $H^1$ data at mass-energy threshold in the inter-range case. In the earlier works of Arora-Roudenko [Arora-Roudenko 2021], the behavior of solutions below the mass-energy threshold was classified. In this paper, we first exhibit three special solutions: $e^{it} Q$, $Q^\pm$, where $Q^\pm$ exponentially approach to the $e^{it} Q$ in the positive time direction, $Q^+$ blows up and $Q^-$ scatters in the negative time direction. Then we classify solutions at this threshold, showing that they behave exactly as the above three special solutions up to symmetries, or scatter or blow up in both time directions. The argument relies on the uniqueness and non-degeneracy of ground state, which we regard as an assumption for the general case.

math.AP↗

arXiv2022

Solving time dependent Fokker-Planck equations via temporal normalizing flow

In this work, we propose an adaptive learning approach based on temporal normalizing flows for solving time-dependent Fokker-Planck (TFP) equations. It is well known that solutions of such equations are probability density functions, and thus our approach relies on modelling the target solutions with the temporal normalizing flows. The temporal normalizing flow is then trained based on the TFP loss function, without requiring any labeled data. Being a machine learning scheme, the proposed approach is mesh-free and can be easily applied to high dimensional problems. We present a variety of test problems to show the effectiveness of the learning approach.

cs.LG↗