arXiv subjects

Jie Wang

Publications and source records attributed to Jie Wang.

At least 397 records · Page 22Linked to original sources

arXiv2023

HiREN: Towards Higher Supervision Quality for Better Scene Text Image Super-Resolution

Scene text image super-resolution (STISR) is an important pre-processing technique for text recognition from low-resolution scene images. Nowadays, various methods have been proposed to extract text-specific information from high-resolution (HR) images to supervise STISR model training. However, due to uncontrollable factors (e.g. shooting equipment, focus, and environment) in manually photographing HR images, the quality of HR images cannot be guaranteed, which unavoidably impacts STISR performance. Observing the quality issue of HR images, in this paper we propose a novel idea to boost STISR by first enhancing the quality of HR images and then using the enhanced HR images as supervision to do STISR. Concretely, we develop a new STISR framework, called High-Resolution ENhancement (HiREN) that consists of two branches and a quality estimation module. The first branch is developed to recover the low-resolution (LR) images, and the other is an HR quality enhancement branch aiming at generating high-quality (HQ) text images based on the HR images to provide more accurate supervision to the LR images. As the degradation from HQ to HR may be diverse, and there is no pixel-level supervision for HQ image generation, we design a kernel-guided enhancement network to handle various degradation, and exploit the feedback from a recognizer and text-level annotations as weak supervision signal to train the HR enhancement branch. Then, a quality estimation module is employed to evaluate the qualities of HQ images, which are used to suppress the erroneous supervision information by weighting the loss of each image. Extensive experiments on TextZoom show that HiREN can work well with most existing STISR methods and significantly boost their performances.

cs.CV↗

arXiv2023

Interpretable Diabetic Retinopathy Diagnosis based on Biomarker Activation Map

Deep learning classifiers provide the most accurate means of automatically diagnosing diabetic retinopathy (DR) based on optical coherence tomography (OCT) and its angiography (OCTA). The power of these models is attributable in part to the inclusion of hidden layers that provide the complexity required to achieve a desired task. However, hidden layers also render algorithm outputs difficult to interpret. Here we introduce a novel biomarker activation map (BAM) framework based on generative adversarial learning that allows clinicians to verify and understand classifiers decision-making. A data set including 456 macular scans were graded as non-referable or referable DR based on current clinical standards. A DR classifier that was used to evaluate our BAM was first trained based on this data set. The BAM generation framework was designed by combing two U-shaped generators to provide meaningful interpretability to this classifier. The main generator was trained to take referable scans as input and produce an output that would be classified by the classifier as non-referable. The BAM is then constructed as the difference image between the output and input of the main generator. To ensure that the BAM only highlights classifier-utilized biomarkers an assistant generator was trained to do the opposite, producing scans that would be classified as referable by the classifier from non-referable scans. The generated BAMs highlighted known pathologic features including nonperfusion area and retinal fluid. A fully interpretable classifier based on these highlights could help clinicians better utilize and verify automated DR diagnosis.

eess.IV↗

arXiv2023

Community Detection Graph Convolutional Network for Overlap-Aware Speaker Diarization

The clustering algorithm plays a crucial role in speaker diarization systems. However, traditional clustering algorithms suffer from the complex distribution of speaker embeddings and lack of digging potential relationships between speakers in a session. We propose a novel graph-based clustering approach called Community Detection Graph Convolutional Network (CDGCN) to improve the performance of the speaker diarization system. The CDGCN-based clustering method consists of graph generation, sub-graph detection, and Graph-based Overlapped Speech Detection (Graph-OSD). Firstly, the graph generation refines the local linkages among speech segments. Secondly the sub-graph detection finds the optimal global partition of the speaker graph. Finally, we view speaker clustering for overlap-aware speaker diarization as an overlapped community detection task and design a Graph-OSD component to output overlap-aware labels. By capturing local and global information, the speaker diarization system with CDGCN clustering outperforms the traditional Clustering-based Speaker Diarization (CSD) systems on the DIHARD III corpus.

eess.AS↗

arXiv2023

Circular motion and chaos bound of a charged particle near charged 4D Einstein-Gauss-Bonnet-AdS black holes

We investigate the circular motion and chaos bound of a charged particle near 4D charged AdS black holes in Einstein-Gauss-Bonnet gravity theory. By means of the Jacobian matrix, the analytical form of the Lyapunov exponent of the charged particle is constructed, which satisfies the upper bound when it is on the event horizon. By further expanding the Lyapunov exponent near the horizon and investigating a 4D charged Einstein-Gauss-Bonnet-AdS black hole with different Gauss-Bonnet coupling constant, we find that it has some specific values to determine whether a violation of chaos bound. Besides, we find that in contrast to the static equilibrium, the circular motion of charged particle can have a larger Lyapunov exponent due to the existence of angular momentum. Moreover, we show that the black hole gets closer to the extremal state as the Gauss-Bonnet coupling constant increases, and the bound is more easily violated. In addition, the range of particle charge that may violate the chaotic bound are found for different Gauss-Bonnet coupling constants. The results show that as the GB coupling parameter increases, the value of particle charge required to satisfy the violation of the chaos bound is even smaller.

gr-qc↗

arXiv2023

LMD: A Learnable Mask Network to Detect Adversarial Examples for Speaker Verification

Although the security of automatic speaker verification (ASV) is seriously threatened by recently emerged adversarial attacks, there have been some countermeasures to alleviate the threat. However, many defense approaches not only require the prior knowledge of the attackers but also possess weak interpretability. To address this issue, in this paper, we propose an attacker-independent and interpretable method, named learnable mask detector (LMD), to separate adversarial examples from the genuine ones. It utilizes score variation as an indicator to detect adversarial examples, where the score variation is the absolute discrepancy between the ASV scores of an original audio recording and its transformed audio synthesized from its masked complex spectrogram. A core component of the score variation detector is to generate the masked spectrogram by a neural network. The neural network needs only genuine examples for training, which makes it an attacker-independent approach. Its interpretability lies that the neural network is trained to minimize the score variation of the targeted ASV, and maximize the number of the masked spectrogram bins of the genuine training examples. Its foundation is based on the observation that, masking out the vast majority of the spectrogram bins with little speaker information will inevitably introduce a large score variation to the adversarial example, and a small score variation to the genuine example. Experimental results with 12 attackers and two representative ASV systems show that our proposed method outperforms five state-of-the-art baselines. The extensive experimental results can also be a benchmark for the detection-based ASV defenses.

eess.AS↗

arXiv2023

Subhalo abundance and satellite spatial distribution in Milky Way-Andromeda-like paired haloes

We study the subhalo and satellite populations in haloes similar to the Milky Way (MW)-Andromeda paired configuration in the Millennium II and P-Millennium simulations. We find subhaloes are $5\%-15\%$ more abundant in paired haloes than their isolated counterparts that have the same halo mass and large-scale environmental density. Paired haloes tend to reside in a more isotropic environment than isolated haloes, the shear tensor of their large-scale tidal field is possibly responsible for this difference. We also study the thickness of the spatial distribution of the top 11 most massive satellite galaxies obtained in the semi-analytic galaxy sample constructed from the Millennium II simulation. Moreover, satellites that have lost their host subhaloes due to the resolution limit of the simulation have been taken into account. As a result, we find that the difference in the distribution of the satellite thickness between isolated and paired haloes is indistinguishable, which suggests that the paired configuration is not responsible for the observed plane of satellites in the Milky Way. The results in this study indicate the paired configuration could bring some nonnegligible effect on the subhalo abundance in the investigation of the Milky Way's satellite problems.

astro-ph.GA↗

arXiv2023

Can LLMs like GPT-4 outperform traditional AI tools in dementia diagnosis? Maybe, but not today

Recent investigations show that large language models (LLMs), specifically GPT-4, not only have remarkable capabilities in common Natural Language Processing (NLP) tasks but also exhibit human-level performance on various professional and academic benchmarks. However, whether GPT-4 can be directly used in practical applications and replace traditional artificial intelligence (AI) tools in specialized domains requires further experimental validation. In this paper, we explore the potential of LLMs such as GPT-4 to outperform traditional AI tools in dementia diagnosis. Comprehensive comparisons between GPT-4 and traditional AI tools are conducted to examine their diagnostic accuracy in a clinical setting. Experimental results on two real clinical datasets show that, although LLMs like GPT-4 demonstrate potential for future advancements in dementia diagnosis, they currently do not surpass the performance of traditional AI tools. The interpretability and faithfulness of GPT-4 are also evaluated by comparison with real doctors. We discuss the limitations of GPT-4 in its current state and propose future research directions to enhance GPT-4 in dementia diagnosis.

cs.CL↗

arXiv2023

Modelling the mass accretion histories of dark matter haloes using a Gamma formalism

We present a physical model of the Mass Accretion Histories (MAH) of haloes in concordance with the {\it observed} cosmic star formation rate density (CSFRD). We model the MAHs of dark matter haloes using a Gamma ($Γ$) functional form: $M_h(T) = \frac{M_0}{f_{0}} \, \times \frac{γ(α_h, ~β_h \times (T-Th))}{Γ(α_h)}$, where $M_0$ is the halo mass at present time, $T$ is time, $α_h$ and $β_h$ are parameters we explore, $f_{0}$ is the percentage of the mass of the halo at z = 0 with respect to the final mass of the halo achieved at $T = \infty$. We use the MAHs of haloes obtained from cosmological simulations and analytical models to constrain our model. $f_{0}$ can be described by a power-law ($f_{0} = 1- c \times M_{0}^{d}$). Haloes with small masses have already on average attained most of their final masses. The average $ $ of haloes in the Universe is $ > 0.95$ pointing to the direction that the cosmic MAH/CSFRD is saturated at our era. The average $<β_{h}>$ parameter (the depletion rate of the available dark matter for halo growth) is related to the dynamical timescales of haloes. The $α$ parameter is a power-law index of $M_{0}$ and represents the early growth a halo experiences before the expansion of the Universe starts to slow it down. Finally, $T_{h}$ (the time that marks the co-evolution/growth of galaxies and haloes after the Big Bang) is found to be 150-300 million years.

astro-ph.GA↗

arXiv2023

Chiral Kondo Lattice in Doped MoTe$_2$/WSe$_2$ Bilayers

We theoretically study the interplay between magnetism and a heavy Fermi liquid in the AB stacked transition metal dichalcogenide bilayer system MoTe$_2$/WSe$_2$ in the regime in which the Mo layer supports localized magnetic moments coupled by interlayer electron tunnelling to a weakly correlated band of itinerant electrons in the W layer. We show that the interlayer electron transfer leads to a chiral Kondo exchange, with consequences including a strong dependence of the Kondo temperature on carrier concentration, a topological hybridization gap and an anomalous Hall effect. The theoretical model exhibits two phases, a small Fermi surface magnet and a large Fermi surface heavy Fermi liquid; the transition between them is first order. A low-energy theory is developed for the the transport properties of the two states. Implications of our results for present and future experiments on MoTe$_2$/WSe$_2$ bilayer heterostructures are discussed.

cond-mat.str-el↗

arXiv2023

A Fast Hybrid Cascade Network for Voxel-based 3D Object Classification

Voxel-based 3D object classification has been thoroughly studied in recent years. Most previous methods convert the classic 2D convolution into a 3D form that will be further applied to objects with binary voxel representation for classification. However, the binary voxel representation is not very effective for 3D convolution in many cases. In this paper, we propose a hybrid cascade architecture for voxel-based 3D object classification. It consists of three stages composed of fully connected and convolutional layers, dealing with easy, moderate, and hard 3D models respectively. Both accuracy and speed can be balanced in our proposed method. By giving each voxel a signed distance value, an obvious gain regarding the accuracy can be observed. Besides, the mean inference time can be speeded up hugely compared with the state-of-the-art point cloud and voxel based methods.

cs.CV↗

arXiv2023

A Manifold Two-Sample Test Study: Integral Probability Metric with Neural Networks

Two-sample tests are important areas aiming to determine whether two collections of observations follow the same distribution or not. We propose two-sample tests based on integral probability metric (IPM) for high-dimensional samples supported on a low-dimensional manifold. We characterize the properties of proposed tests with respect to the number of samples $n$ and the structure of the manifold with intrinsic dimension $d$. When an atlas is given, we propose two-step test to identify the difference between general distributions, which achieves the type-II risk in the order of $n^{-1/\max\{d,2\}}$. When an atlas is not given, we propose Hölder IPM test that applies for data distributions with $(s,β)$-Hölder densities, which achieves the type-II risk in the order of $n^{-(s+β)/d}$. To mitigate the heavy computation burden of evaluating the Hölder IPM, we approximate the Hölder function class using neural networks. Based on the approximation theory of neural networks, we show that the neural network IPM test has the type-II risk in the order of $n^{-(s+β)/d}$, which is in the same order of the type-II risk as the Hölder IPM test. Our proposed tests are adaptive to low-dimensional geometric structure because their performance crucially depends on the intrinsic dimension instead of the data dimension.

stat.ML↗

arXiv2023

AI-assisted reconstruction of cosmic velocity field from redshift-space spatial distribution of halos

The peculiar velocities of dark matter halos are crucial to study many issues in cosmology and galaxy evolution. In this study, by using the state-of-the-art deep learning technique, a UNet-based neural network, we propose to reconstruct the peculiar velocity field from the redshift-space distribution of dark matter halos. Through a point-to-point comparison and examination of various statistical properties, we demonstrate that, the reconstructed velocity field is in good agreement with the ground truth. The power spectra of various velocity field components, including velocity magnitude, divergence and vorticity, can be successfully recovered when $k\lesssim 1.1$ $h/\rm Mpc$ (the Nyquist frequency of the simulations) at about 80% accuracy. This approach is very promising and presents an alternative method to correct the redshift-space distortions using the measured 3D spatial information of halos. Additionally, for the reconstruction of the momentum field of halos, UNet achieves similar good results. Hence the applications in various aspects of cosmology are very broad, such as correcting redshift errors and improving measurements in the structure of the cosmic web, the kinetic Sunyaev-Zel'dovich effect, BAO reconstruction, etc.

astro-ph.CO↗

arXiv2023

Pressure--enhanced fractional Chern insulators in moiré transition metal dichalcogenides along a magic line

We show that pressure applied to twisted WSe$_2$ can enhance the many-body gap and region of stability of a fractional Chern insulator at filling $ν= 1/3$. Our results are based on exact diagonalization of a continuum model, whose pressure-dependence is obtained through {\it ab initio} methods. We interpret our results in terms of a {\it magic line} in the pressure-{\it vs}-twist angle phase diagram: along the magic line, the bandwidth of the topmost moiré valence band is minimized while simultaneously its quantum geometry nearly resembles that of an ideal Chern band. We expect our results to generalize to other twisted transition metal dichalcogenide homobilayers.

cond-mat.str-el↗

arXiv2023

FraudAuditor: A Visual Analytics Approach for Collusive Fraud in Health Insurance

Collusive fraud, in which multiple fraudsters collude to defraud health insurance funds, threatens the operation of the healthcare system. However, existing statistical and machine learning-based methods have limited ability to detect fraud in the scenario of health insurance due to the high similarity of fraudulent behaviors to normal medical visits and the lack of labeled data. To ensure the accuracy of the detection results, expert knowledge needs to be integrated with the fraud detection process. By working closely with health insurance audit experts, we propose FraudAuditor, a three-stage visual analytics approach to collusive fraud detection in health insurance. Specifically, we first allow users to interactively construct a co-visit network to holistically model the visit relationships of different patients. Second, an improved community detection algorithm that considers the strength of fraud likelihood is designed to detect suspicious fraudulent groups. Finally, through our visual interface, users can compare, investigate, and verify suspicious patient behavior with tailored visualizations that support different time scales. We conducted case studies in a real-world healthcare scenario, i.e., to help locate the actual fraud group and exclude the false positive group. The results and expert feedback proved the effectiveness and usability of the approach.

cs.HC↗

arXiv2023

RAF: Holistic Compilation for Deep Learning Model Training

As deep learning is pervasive in modern applications, many deep learning frameworks are presented for deep learning practitioners to develop and train DNN models rapidly. Meanwhile, as training large deep learning models becomes a trend in recent years, the training throughput and memory footprint are getting crucial. Accordingly, optimizing training workloads with compiler optimizations is inevitable and getting more and more attentions. However, existing deep learning compilers (DLCs) mainly target inference and do not incorporate holistic optimizations, such as automatic differentiation and automatic mixed precision, in training workloads. In this paper, we present RAF, a deep learning compiler for training. Unlike existing DLCs, RAF accepts a forward model and in-house generates a training graph. Accordingly, RAF is able to systematically consolidate graph optimizations for performance, memory and distributed training. In addition, to catch up to the state-of-the-art performance with hand-crafted kernel libraries as well as tensor compilers, RAF proposes an operator dialect mechanism to seamlessly integrate all possible kernel implementations. We demonstrate that by in-house training graph generation and operator dialect mechanism, we are able to perform holistic optimizations and achieve either better training throughput or larger batch size against PyTorch (eager and torchscript mode), XLA, and DeepSpeed for popular transformer models on GPUs.

cs.LG↗

arXiv2023

De Novo Molecular Generation via Connection-aware Motif Mining

De novo molecular generation is an essential task for science discovery. Recently, fragment-based deep generative models have attracted much research attention due to their flexibility in generating novel molecules based on existing molecule fragments. However, the motif vocabulary, i.e., the collection of frequent fragments, is usually built upon heuristic rules, which brings difficulties to capturing common substructures from large amounts of molecules. In this work, we propose a new method, MiCaM, to generate molecules based on mined connection-aware motifs. Specifically, it leverages a data-driven algorithm to automatically discover motifs from a molecule library by iteratively merging subgraphs based on their frequency. The obtained motif vocabulary consists of not only molecular motifs (i.e., the frequent fragments), but also their connection information, indicating how the motifs are connected with each other. Based on the mined connection-aware motifs, MiCaM builds a connection-aware generator, which simultaneously picks up motifs and determines how they are connected. We test our method on distribution-learning benchmarks (i.e., generating novel molecules to resemble the distribution of a given training set) and goal-directed benchmarks (i.e., generating molecules with target properties), and achieve significant improvements over previous fragment-based baselines. Furthermore, we demonstrate that our method can effectively mine domain-specific motifs for different tasks.

cs.LG↗

arXiv2023

Sparse Bayesian Learning-Based 3D Spectrum Environment Map Construction-Sampling Optimization, Scenario-Dependent Dictionary Construction and Sparse Recovery

The spectrum environment map (SEM), which can visualize the information of invisible electromagnetic spectrum, is vital for monitoring, management, and security of spectrum resources in cognitive radio (CR) networks. In view of a limited number of spectrum sensors and constrained sampling time, this paper presents a new three-dimensional (3D) SEM construction scheme based on sparse Bayesian learning (SBL). Firstly, we construct a scenario-dependent channel dictionary matrix by considering the propagation characteristic of the interested scenario. To improve sampling efficiency, a maximum mutual information (MMI)-based optimization algorithm is developed for the layout of sampling sensors. Then, a maximum and minimum distance (MMD) clustering-based SBL algorithm is proposed to recover the spectrum data at the unsampled positions and construct the whole 3D SEM. We finally use the simulation data of the campus scenario to construct the 3D SEMs and compare the proposed method with the state-of-the-art. The recovery performance and the impact of different sparsity on the constructed SEMs are also analyzed. Numerical results show that the proposed scheme can reduce the required spectrum sensor number and has higher accuracy under the low sampling rate.

eess.SP↗

arXiv2023

Robust Representation Learning by Clustering with Bisimulation Metrics for Visual Reinforcement Learning with Distractions

Recent work has shown that representation learning plays a critical role in sample-efficient reinforcement learning (RL) from pixels. Unfortunately, in real-world scenarios, representation learning is usually fragile to task-irrelevant distractions such as variations in background or viewpoint. To tackle this problem, we propose a novel clustering-based approach, namely Clustering with Bisimulation Metrics (CBM), which learns robust representations by grouping visual observations in the latent space. Specifically, CBM alternates between two steps: (1) grouping observations by measuring their bisimulation distances to the learned prototypes; (2) learning a set of prototypes according to the current cluster assignments. Computing cluster assignments with bisimulation metrics enables CBM to capture task-relevant information, as bisimulation metrics quantify the behavioral similarity between observations. Moreover, CBM encourages the consistency of representations within each group, which facilitates filtering out task-irrelevant information and thus induces robust representations against distractions. An appealing feature is that CBM can achieve sample-efficient representation learning even if multiple distractions exist simultaneously.Experiments demonstrate that CBM significantly improves the sample efficiency of popular visual RL algorithms and achieves state-of-the-art performance on both multiple and single distraction settings. The code is available at https://github.com/MIRALab-USTC/RL-CBM.

cs.LG↗