检索结果-内蒙古大学图书馆

Pattern Recognition and Machine Learning (PRML), IEEE International Conference on

作者： Jiahui Lv Jiaxin Yang Jun Lei Chao Chen Hong Duan Shuohao Li Laboratory for Big Data and Decision National University of Defense Technology Changsha Hunan National University of Defense Technology Changsha Hunan

ISBN: (数字)9798350355925

ISBN: (纸本)9798350355932

In addressing the inherent challenge of imbalanced data distribution within long-tailed classification problems and the limitations of traditional classification models in fully extracting intricate information from input images, this paper introduces a pioneering approach: the Tail-Vigor Network (TVNet). Specifically tailored for long-tailed image classification tasks, TVNet aims to explore rare patterns, enhance features, and reveal hidden information. To achieve this, a novel multi-branch network is incorporated within the ResNet architecture to extract and integrate information from various hierarchical levels. Furthermore, this approach involves integrating primitive features from different hierarchical levels into an Adaptive-Context Module (ACM) for feature enhancement, which combines adaptive receptive fields with cross-scale contextual attention mechanisms. Additionally, a novel ReLU activation function is custom-designed to address challenges posed by long-tailed data distributions as it encompasses both linear response stages and nonlinear saturation stages. Empirical validation across multiple public datasets demonstrates that the proposed model yields noteworthy performance enhancements in the realm of long-tailed classification tasks, thus substantiating the efficacy of the proposed approach.

关键词： Training Technological innovation Heavily-tailed distribution Attention mechanisms Benchmark testing Feature extraction data models Complexity theory data mining Image classification

来源：评论

学校读者我要写书评

暂无评论

Backtracing Social Events of Interest via Logical Correlation Using GDELT 9

Backtracing Social Events of Interest via Logical Correlatio...

引用

9th International Conference on big data and Information Analytics, bigDIA 2023

作者： Zhao, Dongxu Zhang, Xin Pan, Yan Shi, Honglin Qian, Liwei Yang, Guang National University of Defense Technology Science and Technology on Information Systems Engineering Laboratory Changsha China National University of Defense Technology Laboratory for Big Data and Decision Changsha China National University of Defense Technology College of Systems Engineering Changsha China

ISBN: (纸本)9798350330076

In response to the inability of mainstream methods in existing social event backtracing studies to address the logical correlation problem, the insufficient and inaccurate utilization of event characteristics for implementing logical association backtracing events, we propose a backtracing method for social events of interest. The basic idea is to obtain a set of related events given a small amount of labeled data on interested events, and to construct the causal chain of event occurrence through their merging and fusion. Specifically, we design a node feature construction strategy to select topic and title information from various elements of events to construct node features, thereby addressing the problem of insufficient and inaccurate node feature composition. We also design an adjacency relationship construction strategy, which uses the co-occurrence of participants in event records as the basis for constructing adjacency relationships, and completes the final construction of adjacency relationships by considering time elements comprehensively, thereby resolving the problem of inaccurate construction of inter-event correlation relationships. We construct a multi-layer graph convolutional network to learn the associated features and achieve node classification through a normalization function, thereby solving the problem of logical association inference between events. Experimental results demonstrate that LC-BTE is effective in backtracing analysis of events related to social events of interest. Additionally, this paper explores the performance changes of the method under semi-supervised mode and conducts experimental analysis. Finally, this paper provides the causal chain of events based on the backtracing results. © 2023 IEEE.

关键词： backtracing GDELT logical correlation social events

来源：评论

学校读者我要写书评

暂无评论

Three-Dimensional Collision Avoidance Method for Robot-Assisted Minimally Invasive Surgery

引用

Cyborg and Bionic Systems 2023年第1期4卷 121-135页

作者： Ling Li Xiaojian Li Bo Ouyang Hangjie Mo Hongliang Ren Shanlin Yang School of Management Hefei University of TechnologyHefeiChina Key Laboratory of Process Optimization and Intelligent Decision-Making(Ministry of Education) Hefei University of TechnologyHefeiChina Philosophy and Social Sciences Laboratory of Data Science and Smart Society Governance(Ministry of Education) Hefei University of TechnologyHefeiChina Department of Biomedical Engineering Faculty of EngineeringNational University of SingaporeSingaporeSingapore Department of Electronic Engineering Shun Hing Institute of Advanced EngineeringThe Chinese University of Hong KongHong KongChina National Engineering Laboratory for Big Data Distribution and Exchange Technologies ShanghaiChina

In the robot-assisted minimally invasive surgery,if a collision occurs,the robot system program could be damaged,and normal tissues could be *** avoid collisions during surgery,a 3-dimensional collision avoidance method is proposed in this *** proposed method is predicated on the design of 3 strategic vectors:the collision-with-instrument-avoidance(CI)vector,the collision-with-tissues-avoidance(CT)vector,and the constrained-control(CC)*** CI vector demarcates 3 specific directions to forestall collision among the surgical *** CT vector,on the other hand,comprises 2 components tailored to prevent inadvertent contact between the robot-controlled instrument and nontarget ***,the CC vector is introduced to guide the endpoint of the robot-controlled instrument toward the desired position,ensuring precision in its movements,in alignment with the surgical *** results verify the proposed collision avoidance method for robot-assisted minimally invasive *** code and data are available at https://***/cynerelee/collision-avoidance.

关键词： collision desired verify

来源：评论

学校读者我要写书评

暂无评论

A structural VAR and VECM modeling method for open-high-low-close data contained in candlestick chart

引用

Financial Innovation 2024年第1期10卷 2017-2045页

作者： Wenyang Huang Huiwen Wang Shanshan Wang College of Economics and Management China Agricultural UniversityBeijingChina School of Economics and Management Beihang UniversityBeijingChina Key Laboratory of Complex System Analysis Management and Decision(Beihang University)Ministry of EducationBeijingChina Beijing Advanced Innovation Center for Big Data and Brain Computing BeijingChina

The structural modeling of open-high-low-close(OHLC)data contained within the candlestick chart is crucial to financial ***,the inherent constraints in OHLC data pose immense challenges to its structural *** that fail to process these constraints may yield results deviating from those of the original OHLC data *** address this issue,a novel unconstrained transformation method,along with its explicit inverse transformation,is proposed to properly handle the inherent constraints of OHLC data.A flexible and effective framework for structurally modeling OHLC data is designed,and the detailed procedure for modeling OHLC data through the vector autoregression and vector error correction model are provided as an example of multivariate time-series *** simulations and three authentic financial datasets from the Kweichow Moutai,CSI 100 index,and 50 ETF of the Chinese stock market demonstrate the effectiveness and stability of the proposed modeling *** modeling results of support vector regression provide further evidence that the proposed unconstrained transformation not only ensures structural forecasting of OHLC data but also is an effective feature-extraction method that can effectively improve the forecasting accuracy of machine-learning models for close prices.

关键词： OHLC data Structural modeling Unconstrained transformation Candlestick chart VAR VECM

来源：评论

学校读者我要写书评

暂无评论

GPR B-scan data Classification through Deep Learning Approach 9

GPR B-scan Data Classification through Deep Learning Approac...

引用

9th International Conference on big data and Information Analytics, bigDIA 2023

作者： Zhang, Yaohong Xiao, Kaiming Liu, Kehui Ge, Bin Chen, Haiwen Li, Donghui National University of Defense Technology Laboratory for Big Data and Decision Changsha China Beijing Academy of Science and Technology Institute of Urban Systems Engineering Beijing China Hunan Agricultural University College of Information and Intelligence Changsha China

ISBN: (纸本)9798350330076

Ground Penetrating Radar (GPR) has proven indispensable in non-destructive geophysical investigations, offering a window into subsurface structures using electromagnetic waves. Its prowess in identifying underground utilities and hazards, like cavities, has been universally recognized. However, the intricate B-scan data generated by GPR poses significant interpretation challenges. Traditional methodologies, especially those reliant on dielectric constants and earlier machine learning models, often necessitate prior subsurface knowledge, inhibiting their universal applicability. Enter deep learning, which, in recent years, has changed the landscape of data interpretation across domains. Among the various architectures, Convolutional Neural Networks (CNNs) have shown immense potential in deciphering GPR data. This paper champions a novel methodology that harnesses the power of the ResNet18 architecture. Uniquely, our approach begins with a model pre-trained on the WiderFace dataset, renowned for facial recognition, and subsequently fine-tuned for GPR data. This merger of facial pattern recognition with geophysical data interpretation creates a synergy that promises unprecedented accuracy in subsurface detection. In summation, our research showcases how deep learning, especially the fusion of facial recognition insights with geophysical data, can redefine the boundaries of GPR data interpretation, offering enhanced accuracy and adaptability over conventional methods. © 2023 IEEE.

关键词： Deep learning

来源：评论

学校读者我要写书评

暂无评论

Kernel-Based Causal Discovery in Large Time Series datasets 8

Kernel-Based Causal Discovery in Large Time Series Datasets

引用

8th International Conference on big data and Information Analytics, bigDIA 2022

作者： Liu, Lihua Li, Yanmin Huang, Hongbin Chen, Qiuju Sun, Jiao Wang, Mao National University of Defense Technology Laboratory for Big Data and Decision Changsha China School of Information Science and Technology University of Science and Technology of China Hefei China

ISBN: (数字)9781665487962

ISBN: (纸本)9781665487962

Discovering causal relationships from a large-scale observational time series dataset is an emerging topic in the current field of research, such as neuroscience and system monitoring. Traditional causal inference approaches in such fields are challenging since the curse of dimensionality and the large sample size. Here, we propose a kernel-based causal discovery (KCD) method by capitalizing on a characterization of conditional independence in reproducing kernel Hilbert spaces. Specifically, KCD uses an iterative and heuristic procedure to add variables into the causal parents set and then adopts the kernel-based conditional independence test to determine the true causal parents. We validated this approach on a large-scale synthetic dataset that simulates the typical properties of real-world data. The results of the experiments have shown that the proposed method is superior to currently existing technologies in terms of detection capabilities, which provides a completely new way to explore how to discover and quantify causal networks from time series in the current research field.a © 2022 IEEE.

关键词： Time series

来源：评论

学校读者我要写书评

暂无评论

Continuous Learning for Small-Sample Entity Extraction Based on Self-Training and data Augmentation

Continuous Learning for Small-Sample Entity Extraction Based...

引用

International Conference on big data and Information Analytics (bigDIA)

作者： Jiwei Yang Fei Wu Chong Zhang Wen Tang Jibing Wu Hongbin Huang Laboratory for Big Data and Decision National University of Defense Technology Changsha China National University of Defense Technology Changsha China

ISBN: (数字)9798350354621

ISBN: (纸本)9798350354638

Entity extraction aims to automatically identify and extract entities with specific meanings from text. It is a fundamental and crucial step in the construction of knowledge graphs, and its performance directly impacts the overall quality of the knowledge graph. Given its significant role in knowledge graph construction, this paper addresses two key issues with traditional entity extraction models: reliance on large amounts of annotated data and inability to continuously learn new emerging entity types. To tackle these problems, we propose a method for small-sample entity extraction under continuous learning based on self-training and data augmentation. This method effectively enhances the performance of small-sample entity extraction models in a continuous learning context.

关键词： Training Accuracy Training data Knowledge graphs big data data augmentation data models Stability analysis data mining Context modeling

来源：评论

学校读者我要写书评

暂无评论

Cross-modal Multi-task Learning for Multimedia Event Extraction 39

Cross-modal Multi-task Learning for Multimedia Event Extract...

引用

39th Annual AAAI Conference on Artificial Intelligence, AAAI 2025

作者： Cao, Jianwei Hu, Yanli Tan, Zhen Zhao, Xiang National Key Laboratory of Information Systems Engineering National University of Defense Technology China Laboratory for Big Data and Decision National University of Defense Technology China

ISBN: (纸本)157735897X

Multimedia event extraction aims to jointly extract event structural knowledge from multiple modalities, thus improving the comprehension and utilization of events in the growing multimedia content (e.g., multimedia news). A key challenge in multimedia event extraction is to establish cross-modal correlations during training without multimedia event annotations. Considering the complexity and cost of annotation across modalities, the multimedia event extraction task only provides parallel annotated data for evaluation. Previous works attempt to learn implicit correlations directly from unlabeled image-text pairs, but do not yield substantially better performance for event-centric tasks. To address this problem, we propose a cross-modal multi-task learning framework X-MTL to establish cross-modal correlations at the task level, which can simultaneously address four key tasks of multimedia event extraction: trigger detection, argument extraction, verb classification, and role classification. Specifically, to process inputs from different modalities and tasks, we utilize two separate modality-specific encoders and a modality-shared encoder to learn joint task representations, and introduce textual and visual prompt learning methods to enrich and unify task inputs. To resolve task conflict in cross-modal multi-task learning, we propose a pseudo label based knowledge distillation method, combined with dynamic weight adjustment method, which can effectively lift the performance to surpass the separately-trained models. On the MultiMedia Event Extraction benchmark M2E2, experimental results show that X-MTL surpasses the current state-of-the-art (SOTA) methods by 4.1% for multimedia event mention and 8.2% for multimedia argument role. Copyright © 2025, Association for the Advancement of Artificial Intelligence (***). All rights reserved.

关键词： Multi-task learning

来源：评论

学校读者我要写书评

暂无评论

Multi-Task Model Fusion with Mixture of Experts Structure 9

Multi-Task Model Fusion with Mixture of Experts Structure

引用

9th International Conference on big data and Information Analytics, bigDIA 2023

作者： Zhou, Jingxuan Xiao, Wenhua Zhang, Dayu Fei, Bowen Zhang, Shuai Chen, Fuxu National University of Defense Technology Laboratory for Big Data and Decision ChangSha China Academy of Military Sciences Beijing China Liaoning Technical University Software College Huludao China National University of Defense Technology ChangSha China

ISBN: (纸本)9798350330076

Model fusion, which aggregates multiple models into a single one, presents significant challenges due to the differences between various task models. To simplify model construction and enhance generalization, we propose a multi-task model construction method that employs a mixture of experts architecture. This method considers each task model as an expert, freezing its network structure, and facilitates model fusion by solely training the gating network and adjusting the output of the expert networks. However, the limited capacity of the gating network complicates the accurate allocation of outputs from each expert. To mitigate this issue, we introduce an auxiliary loss function to guide the gating network in aligning inputs with their corresponding expert networks. We validated our approach on the CIFAR-10 dataset, and the experimental results demonstrate the advantages of our method in terms of overall model accuracy, convergence efficiency, and training accuracy. © 2023 IEEE.

关键词： Mixture of Experts Model Fusion Multi-task

来源：评论

学校读者我要写书评

暂无评论

Learning Accurate, Efficient, and Interpretable MLPs on Multiplex Graphs via Node-wise Multi-View Ensemble Distillation

arXiv

引用

arXiv 2025年

作者： Liu, Yunhui Tao, Zhen Zhao, Xiang Zhao, Jianhua Zheng, Tao He, Tieke State Key Laboratory for Novel Software Technology Nanjing University Nanjing China Laboratory for Big Data and Decision National University of Defense Technology Hunan China

Multiplex graphs, with multiple edge types (graph views) among common nodes, provide richer structural semantics and better modeling capabilities. Multiplex Graph Neural Networks (MGNNs), typically comprising view-specific GNNs and a multi-view integration layer, have achieved advanced performance in various downstream tasks. However, their reliance on neighborhood aggregation poses challenges for deployment in latency-sensitive applications. Motivated by recent GNN-to-MLP knowledge distillation frameworks, we propose Multiplex Graph-Free Neural Networks (MGFNN and MGFNN+) to combine MGNNs’ superior performance and MLPs’ efficient inference via knowledge distillation. MGFNN directly trains student MLPs with node features as input and soft labels from teacher MGNNs as targets. MGFNN+ further employs a low-rank approximation-based reparameterization to learn node-wise coefficients, enabling adaptive knowledge ensemble from each view-specific GNN. This node-wise multi-view ensemble distillation strategy allows student MLPs to learn more informative multiplex semantic knowledge for different nodes. Experiments show that MGFNNs achieve average accuracy improvements of about 10% over vanilla MLPs and perform comparably or even better to teacher MGNNs (accurate);MGFNNs achieve a 35.40×-89.14× speedup in inference over MGNNs (efficient);MGFNN+ adaptively assigns different coefficients for multi-view ensemble distillation regarding different nodes (interpretable). Copyright © 2025, The Authors. All rights reserved.

关键词： Graph neural networks

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：