arXiv subjects

Chunhui Li

Publications and source records attributed to Chunhui Li.

35 records · Page 2Linked to original sources

arXiv2024

Distributed Identification of Stable Large-Scale Isomorphic Nonlinear Networks Using Partial Observations

Distributed parameter identification for large-scale multi-agent networks encounters challenges due to nonlinear dynamics and partial observations. Simultaneously, ensuring the stability is crucial for the robust identification of dynamic networks, especially under data and model uncertainties. To handle these challenges, this paper proposes a particle consensus-based expectation maximization (EM) algorithm. The E-step proposes a distributed particle filtering approach, using local observations from agents to yield global consensus state estimates. The M-step constructs a likelihood function with an a priori contraction-stabilization constraint for the parameter estimation of isomorphic agents. Performance analysis and simulation results of the proposed method confirm its effectiveness in identifying parameters for stable nonlinear networks.

math.DS↗

arXiv2023

CPIA Dataset: A Comprehensive Pathological Image Analysis Dataset for Self-supervised Learning Pre-training

Pathological image analysis is a crucial field in computer-aided diagnosis, where deep learning is widely applied. Transfer learning using pre-trained models initialized on natural images has effectively improved the downstream pathological performance. However, the lack of sophisticated domain-specific pathological initialization hinders their potential. Self-supervised learning (SSL) enables pre-training without sample-level labels, which has great potential to overcome the challenge of expensive annotations. Thus, studies focusing on pathological SSL pre-training call for a comprehensive and standardized dataset, similar to the ImageNet in computer vision. This paper presents the comprehensive pathological image analysis (CPIA) dataset, a large-scale SSL pre-training dataset combining 103 open-source datasets with extensive standardization. The CPIA dataset contains 21,427,877 standardized images, covering over 48 organs/tissues and about 100 kinds of diseases, which includes two main data types: whole slide images (WSIs) and characteristic regions of interest (ROIs). A four-scale WSI standardization process is proposed based on the uniform resolution in microns per pixel (MPP), while the ROIs are divided into three scales artificially. This multi-scale dataset is built with the diagnosis habits under the supervision of experienced senior pathologists. The CPIA dataset facilitates a comprehensive pathological understanding and enables pattern discovery explorations. Additionally, to launch the CPIA dataset, several state-of-the-art (SOTA) baselines of SSL pre-training and downstream evaluation are specially conducted. The CPIA dataset along with baselines is available at https://github.com/zhanglab2021/CPIA_Dataset.

eess.IV↗

arXiv2023

Unveiling nontrivial fusion rule of Majorana zero mode using a fermionic mode

Fusing Majorana zero modes leads to multiple outcomes, a property being unique to non-Abelian anyons. Successful demonstration of this nontrivial fusion rule would be a hallmark for the development of topological quantum computation.Here we show that this can be done by simply attaching a fermionic mode to a single Majorana zero mode. Through modulation of the energy level of this fermionic mode as well as its coupling with the Majorana mode in different sequences, we show that a zero or integer charge pumping can be realized when different fusion loops are chosen. Such fusion loops are intimately related with the nontrivial fusion rule of Majorana modes and are solely determined by the crossings at zero energy in the parameter space. Finally we demonstrate our proposal in a nanowire-based topological superconductor coupled to a quantum dot. We show that the charge pumping is robust for MZMs in the real system irrespective of the initial condition of FM state, contrary to the case for trivial Andreev bound states. This provides a feasible way to distinguish Majorana modes from trivial Andreev bound states.

cond-mat.mes-hall↗

arXiv2023

M2DF: Multi-grained Multi-curriculum Denoising Framework for Multimodal Aspect-based Sentiment Analysis

Multimodal Aspect-based Sentiment Analysis (MABSA) is a fine-grained Sentiment Analysis task, which has attracted growing research interests recently. Existing work mainly utilizes image information to improve the performance of MABSA task. However, most of the studies overestimate the importance of images since there are many noise images unrelated to the text in the dataset, which will have a negative impact on model learning. Although some work attempts to filter low-quality noise images by setting thresholds, relying on thresholds will inevitably filter out a lot of useful image information. Therefore, in this work, we focus on whether the negative impact of noisy images can be reduced without modifying the data. To achieve this goal, we borrow the idea of Curriculum Learning and propose a Multi-grained Multi-curriculum Denoising Framework (M2DF), which can achieve denoising by adjusting the order of training data. Extensive experimental results show that our framework consistently outperforms state-of-the-art work on three sub-tasks of MABSA.

cs.CL↗

arXiv2023

DRIN: Dynamic Relation Interactive Network for Multimodal Entity Linking

Multimodal Entity Linking (MEL) is a task that aims to link ambiguous mentions within multimodal contexts to referential entities in a multimodal knowledge base. Recent methods for MEL adopt a common framework: they first interact and fuse the text and image to obtain representations of the mention and entity respectively, and then compute the similarity between them to predict the correct entity. However, these methods still suffer from two limitations: first, as they fuse the features of text and image before matching, they cannot fully exploit the fine-grained alignment relations between the mention and entity. Second, their alignment is static, leading to low performance when dealing with complex and diverse data. To address these issues, we propose a novel framework called Dynamic Relation Interactive Network (DRIN) for MEL tasks. DRIN explicitly models four different types of alignment between a mention and entity and builds a dynamic Graph Convolutional Network (GCN) to dynamically select the corresponding alignment relations for different input samples. Experiments on two datasets show that DRIN outperforms state-of-the-art methods by a large margin, demonstrating the effectiveness of our approach.

cs.CL↗

arXiv2023

CellMix: A General Instance Relationship based Method for Data Augmentation Towards Pathology Image Classification

In pathology image analysis, obtaining and maintaining high-quality annotated samples is an extremely labor-intensive task. To overcome this challenge, mixing-based methods have emerged as effective alternatives to traditional preprocessing data augmentation techniques. Nonetheless, these methods fail to fully consider the unique features of pathology images, such as local specificity, global distribution, and inner/outer-sample instance relationships. To better comprehend these characteristics and create valuable pseudo samples, we propose the CellMix framework, which employs a novel distribution-oriented in-place shuffle approach. By dividing images into patches based on the granularity of pathology instances and shuffling them within the same batch, the absolute relationships between instances can be effectively preserved when generating new samples. Moreover, we develop a curriculum learning-inspired, loss-driven strategy to handle perturbations and distribution-related noise during training, enabling the model to adaptively fit the augmented data. Our experiments in pathology image classification tasks demonstrate state-of-the-art (SOTA) performance on 7 distinct datasets. This innovative instance relationship-centered method has the potential to inform general data augmentation approaches for pathology image classification. The associated codes are available at https://github.com/sagizty/CellMix.

cs.CV↗

arXiv2023

Beyond Black Box AI-Generated Plagiarism Detection: From Sentence to Document Level

The increasing reliance on large language models (LLMs) in academic writing has led to a rise in plagiarism. Existing AI-generated text classifiers have limited accuracy and often produce false positives. We propose a novel approach using natural language processing (NLP) techniques, offering quantifiable metrics at both sentence and document levels for easier interpretation by human evaluators. Our method employs a multi-faceted approach, generating multiple paraphrased versions of a given question and inputting them into the LLM to generate answers. By using a contrastive loss function based on cosine similarity, we match generated sentences with those from the student's response. Our approach achieves up to 94% accuracy in classifying human and AI text, providing a robust and adaptable solution for plagiarism detection in academic settings. This method improves with LLM advancements, reducing the need for new model training or reconfiguration, and offers a more transparent way of evaluating and detecting AI-generated text.

cs.CL↗

arXiv2023

Deep-Learning-based Vasculature Extraction for Single-Scan Optical Coherence Tomography Angiography

Optical coherence tomography angiography (OCTA) is a non-invasive imaging modality that extends the functionality of OCT by extracting moving red blood cell signals from surrounding static biological tissues. OCTA has emerged as a valuable tool for analyzing skin microvasculature, enabling more accurate diagnosis and treatment monitoring. Most existing OCTA extraction algorithms, such as speckle variance (SV)- and eigen-decomposition (ED)-OCTA, implement a larger number of repeated (NR) OCT scans at the same position to produce high-quality angiography images. However, a higher NR requires a longer data acquisition time, leading to more unpredictable motion artifacts. In this study, we propose a vasculature extraction pipeline that uses only one-repeated OCT scan to generate OCTA images. The pipeline is based on the proposed Vasculature Extraction Transformer (VET), which leverages convolutional projection to better learn the spatial relationships between image patches. In comparison to OCTA images obtained via the SV-OCTA (PSNR: 17.809) and ED-OCTA (PSNR: 18.049) using four-repeated OCT scans, OCTA images extracted by VET exhibit moderate quality (PSNR: 17.515) and higher image contrast while reducing the required data acquisition time from ~8 s to ~2 s. Based on visual observations, the proposed VET outperforms SV and ED algorithms when using neck and face OCTA data in areas that are challenging to scan. This study represents that the VET has the capacity to extract vascularture images from a fast one-repeated OCT scan, facilitating accurate diagnosis for patients.

eess.IV↗

arXiv2023

EPCS: Endpoint-based Part-aware Curve Skeleton Extraction for Low-quality Point Clouds

The curve skeleton is an important shape descriptor that has been utilized in various applications in computer graphics, machine vision, and artificial intelligence. In this study, the endpoint-based part-aware curve skeleton (EPCS) extraction method for low-quality point clouds is proposed. The novel random center shift (RCS) method is first proposed for detecting the endpoints on point clouds. The endpoints are used as the initial seed points for dividing each part into layers, and then the skeletal points are obtained by computing the center points of the oriented bounding box (OBB) of the layers. Subsequently, the skeletal points are connected, thus forming the branches. Furthermore, the multi-vector momentum-driven (MVMD) method is also proposed for locating the junction points that connect the branches. Due to the shape differences between different parts on point clouds, the global topology of the skeleton is finally optimized by removing the redundant junction points, re-connecting some branches using the proposed MVMD method, and applying an interpolation method based on the splitting operator. Consequently, a complete and smooth curve skeleton is achieved. The proposed EPCS method is compared with several state-of-the-art methods, and the experimental results verify its robustness, effectiveness, and efficiency. Furthermore, the skeleton extraction and model segmentation results on the point clouds of broken Terracotta also highlight the utility of the proposed method.

cs.CV↗

arXiv2022

An Unsupervised Learning Approach for Spectrum Allocation in Terahertz Communication Systems

We propose a new spectrum allocation strategy, aided by unsupervised learning, for multiuser terahertz communication systems. In this strategy, adaptive sub-band bandwidth is considered such that the spectrum of interest can be divided into sub-bands with unequal bandwidths. This strategy reduces the variation in molecular absorption loss among the users, leading to the improved data rate performance. We first formulate an optimization problem to determine the optimal sub-band bandwidth and transmit power, and then propose the unsupervised learning-based approach to obtaining the near-optimal solution to this problem. In the proposed approach, we first train a deep neural network (DNN) while utilizing a loss function that is inspired by the Lagrangian of the formulated problem. Then using the trained DNN, we approximate the near-optimal solutions. Numerical results demonstrate that comparing to existing approaches, our proposed unsupervised learning-based approach achieves a higher data rate, especially when the molecular absorption coefficient within the spectrum of interest varies in a highly non-linear manner.

cs.LG↗

arXiv2022

Truncated Channel Inversion Power Control to Enable One-Way URLLC with Imperfect Channel Reciprocity

We propose to use channel inversion power control (CIPC) to achieve one-way ultra-reliable and low-latency communications (URLLC), where only the transmission in one direction requires ultra reliability and low latency. Based on channel reciprocity, our proposed CIPC schemes guarantee the power of received signal that is used to decode the information to be a constant value $Q$, by varying the transmit signal and power, which relaxes the assumption of knowing channel state information (CSI) at the user. Thus, the CIPC schemes eliminate the overhead of CSI feedback, reduce communication latency, and explore the benefits of multiple antennas to significantly improve transmission reliability. We derive analytical expressions for the packet loss probability of the proposed CIPC schemes, based on which we determine a closed interval and a convex set for optimizing $Q$ in CIPC with imperfect and perfect channel reciprocity, respectively. Our results show that CIPC is an effective means to achieve one-way URLLC. The tradeoff among reliability, latency, and required resources (e.g., transmit antennas) is further revealed, which provides novel principles for designing one-way URLLC systems.

cs.IT↗

arXiv2022

Machine Learning in Heterogeneous Porous Materials

The "Workshop on Machine learning in heterogeneous porous materials" brought together international scientific communities of applied mathematics, porous media, and material sciences with experts in the areas of heterogeneous materials, machine learning (ML) and applied mathematics to identify how ML can advance materials research. Within the scope of ML and materials research, the goal of the workshop was to discuss the state-of-the-art in each community, promote crosstalk and accelerate multi-disciplinary collaborative research, and identify challenges and opportunities. As the end result, four topic areas were identified: ML in predicting materials properties, and discovery and design of novel materials, ML in porous and fractured media and time-dependent phenomena, Multi-scale modeling in heterogeneous porous materials via ML, and Discovery of materials constitutive laws and new governing equations. This workshop was part of the AmeriMech Symposium series sponsored by the National Academies of Sciences, Engineering and Medicine and the U.S. National Committee on Theoretical and Applied Mechanics.

cs.LG↗

arXiv2021

Modeling the light curve of Type IIn-P SN 2005cl with red supergiant progenitors featuring pre-SN ourbursts

All Type IIn supernovae (SNe IIn) show narrow hydrogen emission lines in their spectra. Apart from this common feature, they demonstrate very broad diversity in brightness, duration, and morphology of their light curves, which indicates that they likely come from a variety of progenitor systems and explosion channels. A particular subset of SNe IIn, the so called SNe IIn-P, exhibit $\sim$100 days plateau phases that are very similar to the ones of the ordinary hydrogen-rich SNe (SNe II). In the past, SNe IIn-P were explained by the models of sub-energetic electron capture explosions surrounded by dense extended winds. In this work, we attempt to explain this class of SNe with standard red supergiant (RSG) progenitors that experience outbursts several month before the final explosion. The outburst energies that show the best agreement between our models and the data ($5\times10^{46}\,{\rm erg}$) fall at the low range of the outburst energies that have been observed for SNe IIn (between few times $10^{46}\,{\rm erg}$ and $10^{49}\,{\rm erg}$). Instead, the inferred explosion energy of SN 2005cl is relatively high ($1-2\times10^{51}\,{\rm erg}$) compared to the explosion energies of the ordinary SNe II. Our models provide alternative explanation of SNe IIn-P to the previously proposed scenarios.

astro-ph.HE↗

arXiv2020

An End-to-End Attack on Text-based CAPTCHAs Based on Cycle-Consistent Generative Adversarial Network

As a widely deployed security scheme, text-based CAPTCHAs have become more and more difficult to resist machine learning-based attacks. So far, many researchers have conducted attacking research on text-based CAPTCHAs deployed by different companies (such as Microsoft, Amazon, and Apple) and achieved certain results.However, most of these attacks have some shortcomings, such as poor portability of attack methods, requiring a series of data preprocessing steps, and relying on large amounts of labeled CAPTCHAs. In this paper, we propose an efficient and simple end-to-end attack method based on cycle-consistent generative adversarial networks. Compared with previous studies, our method greatly reduces the cost of data labeling. In addition, this method has high portability. It can attack common text-based CAPTCHA schemes only by modifying a few configuration parameters, which makes the attack easier. Firstly, we train CAPTCHA synthesizers based on the cycle-GAN to generate some fake samples. Basic recognizers based on the convolutional recurrent neural network are trained with the fake data. Subsequently, an active transfer learning method is employed to optimize the basic recognizer utilizing tiny amounts of labeled real-world CAPTCHA samples. Our approach efficiently cracked the CAPTCHA schemes deployed by 10 popular websites, indicating that our attack is likely very general. Additionally, we analyzed the current most popular anti-recognition mechanisms. The results show that the combination of more anti-recognition mechanisms can improve the security of CAPTCHA, but the improvement is limited. Conversely, generating more complex CAPTCHAs may cost more resources and reduce the availability of CAPTCHAs.

cs.CV↗

arXiv2019

One-Way URLLC with Truncated Channel Inversion Power Control

In this work, we consider one-way ultra-reliable and low-latency communication (URLLC), where only the transmission in one direction requires URLLC and the transmission in the opposite direction does not. In order to meet the low-latency requirement of the one-way URLLC, we propose to use a truncated channel inversion power control (CIPC) to eliminate the requirement and the associated overhead of the training-based channel estimation at the receiver, while utilizing the multi-antenna technique at the transmitter to enhance the communication reliability. We first derive the transmission outage probability achieved by the truncated CIPC by considering the impact of a finite blocklength and a maximum transmit power constraint. Then, we determine the optimal constant power of the received signals in the truncated CIPC, which minimizes the transmission outage probability. Our examination shows that the proposed truncated CIPC is an effective means to achieve the one-way URLLC, where the tradeoff among reliability, latency, and required resources (e.g., the required number of transmit antennas, or the required maximum transmit power) is revealed.

cs.IT↗

arXiv2019

Physical Layer Security for Ultra-Reliable and Low-Latency Communications

Ultra-reliable and low-latency communication (URLLC) is one category of service to be provided by next-generation wireless networks. Motivated by increasing security concerns in such networks, this article focuses on physical layer security (PLS) in the context of URLLC. The PLS technique mainly uses transmission designs based on the intrinsic randomness of the wireless medium to achieve secrecy. As such, PLS is of lower complexity and incurs less latency than traditional cryptography. In this article, we first introduce appropriate performance metrics for evaluating PLS in URLLC, illustrating the tradeoff between latency, reliability, and security. We then identify the key challenging problems for achieving PLS for URLLC, and discuss the role that channel state information can have in providing potential solutions to these problems. Finally, we present our recommendations on future research directions in this emerging area.

cs.IT↗

arXiv2018

On Channel Reciprocity to Activate Uplink Channel Training for Downlink Wireless Transmission in Tactile Internet Applications

We determine, for the first time, the requirement on channel reciprocity to activate uplink channel training, instead of downlink channel training, to achieve a higher data rate for the downlink transmission from a multi-antenna base station to a single-antenna user. We first derive novel closed-form expressions for the lower bounds on the data rates achieved by the two channel training strategies by considering the impact of finite blocklength. The performance comparison result of these two strategies is determined by the amount of channel reciprocity that is utilized in the uplink channel training. We then derive an approximated expression for the minimum channel reciprocity that enables the uplink channel training to outperform the downlink channel training. Through numerical results, we demonstrate that this minimum channel reciprocity decreases as the blocklength decreases or the number of transmit antennas increases, which shows the necessity and benefits of activating the uplink channel training for short-packet communications with multiple transmit antennas. This work provides pivotal and unprecedented guidelines on choosing channel training strategies and channel reciprocity calibrations, offering valuable insights into latency reduction in the Tactile Internet applications.

cs.IT↗