检索结果-内蒙古大学图书馆

SSRN 2024年

作者： Li, Jiaxing Zhang, Xiaotong Zhang, Ke Kong, Youyong Jiangsu Provincial Joint International Research Laboratory of Medical Information Processing School of Computer Science and Engineering Southeast University Jiangsu Province Nanjing China Key Laboratory of Computer Network and Information Integration Southeast University Ministry of Education Nanjing China

Graph neural networks (GNNs) are powerful architectures for graph representation learning. However, GNNs based on message-passing cannot make fully use of the valuable long-range dependencies in graph-structured data. Recently, researchers have designed Transformer-based graph learning models to expand the receptive field to capture long-range dependencies between pairwise nodes. Although the preliminary performance of the graph Transformer is encouraging and even surpasses GNNs, the quadratic complexity of the self-attention leads to much longer inference time than GNNs, which limits the deployment in practical applications. To address this issue, we propose a knowledge distillation framework to transfer long-range dependencies captured by graph Transformer to GNNs. Our framework extracts knowledge from a pre-trained graph Transformer teacher and transfer it to a GNN-based student. To better transfer the long-range dependencies, we further design a long-range dependencies transfer module, in which the intermediate graph representations of the graph Transformer and GNNs are extracted as distributions. Then we minimize the distance between the two distributions, thereby achieving the long-range dependencies transfer more accurately. The proposed method can be extended to any graph Transformer and GNNs. We evaluate our proposed method on a large number of graph classification benchmark datasets with different GNN-based baselines, and the experimental results demonstrate that our method can obtain the performance close to graph Transformer with faster inference speed and smaller memory usage. © 2024, The Authors. All rights reserved.

关键词： Distillation

来源：评论

学校读者我要写书评

暂无评论

Offensive Text Style Transfer based on Unsupervised Learning

Offensive Text Style Transfer based on Unsupervised Learning

引用

2022 International Conference on Artificial Intelligence and Intelligent information Processing, AIIIP 2022

作者： Zhao, Hanlin Yang, Peng School of Cyber Science and Engineering Southeast University Nanjing China School of Computer Science and Engineering Southeast University Nanjing China Key Laboratory of Computer Network and Information Integration Southeast University Ministry of Education Nanjing China

ISBN: (纸本)9781510660175

As the current mainstream social networking method, online social networking brings convenience to people, but also generates negative problems such as language violence. Therefore, offensive text style transfer research has become an arduous task. Research has shown that the semantic loss is caused by the discreteness and timing of the text data, resulting in a low degree of content retention after style transferring, and the lack of relevant parallel corpus data and specific style tag keywords, therefore, the accuracy of text style transfer is limited. Aiming at the existing problems, this paper proposes an Offensive Text Style Transfer based on the Unsupervised Learning (OTST-UL) model. First, the input text is encoded through the bidirectional encoding attention mechanism to retain the core content of the text. Then the generator-discriminator is applied for adversarial training, and a reconstruction loss algorithm is constructed to ensure the accuracy of the offensive language style transformation and the integrity of the text content. Experimental results show that the OTST-UL model outperforms existing text style transfer models on offensive language datasets. © 2022 SPIE.

关键词： Unsupervised learning

来源：评论

学校读者我要写书评

暂无评论

Efficient survivable mapping algorithm for logical topology in IP-over-WDM optical networks against node failure

引用

Journal of Supercomputing 2023年第5期79卷 5037-5063页

作者： Cheng, Dun-Wei Chang, Jo-Yi Lin, Chen-Yen Lin, Limei Huang, Yanze Thulasiraman, Krishnaiyan Hsieh, Sun-Yuan Department of Computer Science and Information Engineering National Yunlin University of Science and Technology 123 University Road Yunlin64002 Taiwan Department of Computer Science and Information Engineering National Cheng Kung University No. 1 University Road Tainan701 Taiwan College of Mathematics and Informatics and the Key Laboratory of Network Security and Cryptology Fujian Normal University Fujian Fuzhou350117 China Fujian Provincial Key Laboratory of Big Data Mining and Applications School of Mathematics and Physics Fujian University of Technology Fujian Fuzhou350118 China Department of Computer Science University of Oklahoma 660 Parrington Oval NormanOK73019 United States Department of Computer Science and Information Engineering National Chi Nan University No. 1 Daxue Rd Nantou County Puli Township54561 Taiwan

The survivable mapping problem in an IP-over-WDM network with a logical graph and physical graph is the problem of finding a mapping of a logical network in a physical layer so that any failure in physical topology does not break the logical topology’s connection. Determining whether a survivable mapping against failure exists is an NP-complete problem, and therefore many heuristic algorithms have been proposed in the literature. In this paper, a heuristic mapping design strategy is proposed to enable the lightpaths to more efficiently endure a node failure. We individually consider the logical topology by building a Hamiltonian cycle and then deal with the mapping problem for disjoint physical paths. Experimental results demonstrate that the proposed algorithm can provide efficient survivable mapping in IP-over-WDM networks. The simulation shows that the survivability of previous methods (SNF and SMART-Node) is around 60–80%, but our proposed algorithm can build survivable mapping in all kinds of situations. © 2022, The Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Nature.

关键词： Heuristic algorithms

来源：评论

学校读者我要写书评

暂无评论

Efficient and Long-Tailed Generalization for Pre-trained Vision-Language Model

arXiv

引用

arXiv 2024年

作者： Shi, Jiang-Xin Zhang, Chi Wei, Tong Li, Yu-Feng National Key Laboratory for Novel Software Technology School of Artificial Intelligence Nanjing University China National Key Laboratory for Novel Software Technology Nanjing University China School of Computer Science and Engineering Key Laboratory of Computer Network and Information Integration of Ministry of Education Southeast University China

Pre-trained vision-language models like CLIP have shown powerful zero-shot inference ability via image-text matching and prove to be strong few-shot learners in various downstream tasks. However, in real-world scenarios, adapting CLIP to downstream tasks may encounter the following challenges: 1) data may exhibit long-tailed data distributions and might not have abundant samples for all the classes;2) There might be emerging tasks with new classes that contain no samples at all. To overcome them, we propose a novel framework to achieve efficient and long-tailed generalization, which can be termed as Candle. During the training process, we propose compensating logit-adjusted loss to encourage large margins of prototypes and alleviate imbalance both within the base classes and between the base and new classes. For efficient adaptation, we treat the CLIP model as a black box and leverage the extracted features to obtain visual and textual prototypes for prediction. To make full use of multi-modal information, we also propose cross-modal attention to enrich the features from both modalities. For effective generalization, we introduce virtual prototypes for new classes to make up for their lack of training images. Candle achieves state-of-the-art performance over extensive experiments on 11 diverse datasets while substantially reducing the training time, demonstrating the superiority of our approach. The source code is available at https://***/shijxcs/Candle. Copyright © 2024, The Authors. All rights reserved.

关键词： Zero-shot learning

来源：评论

学校读者我要写书评

暂无评论

ENS-RFMC: An Encrypted network Traffic Sampling Method Based on Rule-Based Feature Extraction and Multi-hierarchical Clustering for Intrusion Detection 27th

ENS-RFMC: An Encrypted Network Traffic Sampling Method Based...

引用

27th International Conference on Pattern Recognition, ICPR 2024

作者： Chen, Liang-Chen Gao, Shu Wei, Zi-Xuan Liu, Bao-Xu Zhang, Xu-Yao School of Computer Science and Artificial Intelligence Wuhan University of Technology Wuhan430063 China Institute of Automation NLPR Chinese Academy of Sciences Beijing100190 China Institute of Information Engineering Chinese Academy of Sciences Beijing100093 China School of Computer China University of Labor Relations Beijing100048 China Key Laboratory of Network Assessment Technology Chinese Academy of Sciences Beijing100093 China

ISBN: (纸本)9783031783821

Efficiently and quickly identifying malware in encrypted network traffic is a major challenge. This research presents and evaluates a novel approach for sampling encrypted network traffic, called ENS-RFMC. Initially, the method employs a rule-based strategy for extracting relevant metadata features of encrypted traffic from network connections, SSL and certificates log files. Subsequently, a multi-hierarchical clustering algorithm is utilized to divide the dataset into smaller clusters, with benign clusters subsequently eliminated for streamline data processing. Classifier models such as Random Forests, XGBoost and SVM are then utilized to detect attacks and assess performance. Experimental results indicate that the proposed ENS-RFMC sampling method is effective, with the encrypted traffic intrusion detection model employing ENS-RFMC sampling exhibits enhanced accuracy and efficiency in identifying attacks. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.

关键词： network intrusion

来源：评论

学校读者我要写书评

暂无评论

Deep Imputation Bi-stochastic Graph Regularized Matrix Factorization for Clustering Single-cell RNA-sequencing Data

引用

IEEE/ACM Transactions on Computational Biology and Bioinformatics 2024年 PP卷 1-11页

作者： Lan, Wei Chen, Jianwei Liu, Mingyang Chen, Qingfeng Liu, Jin Wang, Jianxin Chen, Yi-Ping Phoebe Wei Lan is School of Computer Electronic and Information and Guangxi Key Laboratory of Multimedia Communications and Network Technology Guangxi University Nanning Guangxi China School of Computer Electronic and Information Guangxi University Nanning Guangxi China School of Computer Electronic and Information and State Key Laboratory for Conservation and Utilization of Subtropical Agro-bioresources Guangxi University Nanning Guangxi China Hunan Provincial Key Lab on Bioinformatics School of Computer Science and Engineering Central South University Changsha Hunan China Department of Computer Science and Information Technology La Trobe University Melbourne Victoria Australia

By generating massive gene transcriptome data and analyzing transcriptomic variations at the cell level, single-cell RNA-sequencing (scRNA-seq) technology has provided new way to explore cellular heterogeneity and functionality. Clustering scRNA-seq data could discover the hidden diversity and complexity of cell populations, which can aid to the identification of the disease mechanisms and biomarkers. In this paper, a novel method (DSINMF) is presented for single cell RNA sequencing data by using deep matrix factorization. Our proposed method comprises four steps: first, the feature selection is utilized to remove irrelevant features. Then, the dropout imputation is used to handle missing value problem. Further, the dimension reduction is employed to preserve data characteristics and reduce noise effects. Finally, the deep matrix factorization with bi-stochastic graph regularization is used to obtain cluster results from scRNA-seq data. We compare DSINMF with other state-of-the-art algorithms on nine datasets and the results show our method outperformances than other methods. IEEE

关键词： RNA

来源：评论

学校读者我要写书评

暂无评论

Data-consistent Unsupervised Diffusion Model for Metal Artifact Reduction

Data-consistent Unsupervised Diffusion Model for Metal Artif...

引用

2023 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2023

作者： Tong, Zhan Wu, Zhan Yang, Yang Mao, Weilong Wang, Shijie Li, Yinsheng Chen, Yang Southeast University Laboratory of Image Science and Technology Nanjing210096 China Southeast University Ministry of Education Key Laboratory of Computer Network and Information Integration Nanjing210096 China Chinese Academy of Sciences Research Center for Medical Artificial Intelligence Shenzhen Institutes of Advanced Technology Shenzhen518055 China School of Computer Science and Engineering Key Lab. of New Generation Artificial Intelligence Technology and Its Interdisciplinary Applications Jiangsu Provincial Joint International Research Laboratory of Medical Information Processing The Laboratory of Image Science and Technology Nanjing210096 China

ISBN: (纸本)9798350337488

Computed Tomography (CT) is an imaging technique widely used in clinical diagnosis. However, high-attenuation metallic implants result in the obstruction of low-energy Xrays and further lead to metal artifacts in the reconstructed CT images. Deep supervised model-based metal artifact reduction(MAR) approaches are limited in clinical applications due to the difficulty in obtaining paired artifact-affected and artifactfree data. Furthermore, these model-based methods lack the consideration of data consistency in the sinogram-domain to perform exact metal trace inpainting. To address these challenges, we propose a Data-consistent unsupErVised diffusiOn model for meTal artifact rEDuction, called DEVOTED-Net. First, DEVOTED-Net leverages prior knowledge to guide the conditional diffusion model for fine-grained metal trace inpainting. Second, an unsupervised MAR framework is designed in the reverse process for the unknown metal traces restoration in the sinogram domain. Third, to further enhance the sinogram-domain data consistency, physics-based consistency constraint loss including conjugateray consistency loss and accumulation-ray consistency loss is designed. Extensive experiments are carried out to verify the performance of our algorithm on the publicly available dataset and clinical experimental dataset. This efficient, accurate, and reliable MAR approach holds great potential in clinics. © 2023 IEEE.

关键词： computed tomography deep unsupervised learning denoising diffusion probabilistic model Metal artifact reduction physics-based consistency constraint

来源：评论

学校读者我要写书评

暂无评论

A Data Security Protection Method for Deep Neural network Model Based on Mobility and Sharing 1

引用

18th International Conference on Green, Pervasive, and Cloud Computing, GPC 2023

作者： Zhao, Xinjian Li, Qianmu Wang, Qi Chen, Shi Li, Tengfei Li, Nianzhe State Grid Jiangsu Information and Telecommunication Company Jiangsu Nanjing210024 China School of Computer Science and Engineering Nanjing University of Science and Technology Jiangsu Nanjing210094 China State Grid Smart Grid Research Institute Co. Ltd. Jiangsu Nanjing210003 China State Grid Key Laboratory of Information and Network Security Jiangsu Nanjing210003 China

ISBN: (数字)9789819998937

ISBN: (纸本)9789819998920

With the rapid development of digital economy, numerous business scenarios, such as smart grid, energy network, intelligent transportation, etc., require the design and distribution of deep neural network (DNN) models, which typically use large amounts of data and computing resources for training. As a result, DNN models are now considered important assets, but at great risk of being stolen and distributed illegally. In response to the new risks and challenges brought by frequent data flow and sharing, watermarks have been introduced to protect the ownership of DNN models. Watermarks can be extracted in a relatively simple manner to declare ownership of a model. However, watermarks are vulnerable to attacks. In this work, we propose a novel label-based black-box watermark model protection algorithm. Inspired by new labels, we design a method to embed watermarks by adding new labels in the model, and to prevent watermarks from being forged, we use encryption algorithms to generate key samples. We conduct experiments on the VGG19 model using the CIFAR-10 dataset. Experimental results show that the method is robust to fine-tuning attacks, pruning attacks. Furthermore, our method does not affect the performance of deep neural network models. This method helps to solve scenarios such as "internal data circulation, external data sharing and multi-party data collaboration". © 2024, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.

关键词： Neural network models

来源：评论

学校读者我要写书评

暂无评论

AI-Enabled Multimedia Semantic Communications

引用

IEEE Communications Magazine 2025年第6期63卷 66-73页

作者： Chenxing Li Yiping Duan Liejun Wang Xiaoming Tao Department of Electronic Engineering Tsinghua University China State Key Laboratory of Space Network China Beijing National Research Center for Information Science and Technology China School of Computer Science and Technology Xinjiang University China

With increasing demand for mobile communication and the development of artificial intelligence, the development of the sixth-generation mobile communication network (6G) will be oriented toward more intelligent and multimodal services. It is necessary to consider the deep integration of information, communication, and data technologies and build a full life cycle resource management framework for computing, data, artificial intelligence models, and communication. To address the requirements of 6G intelligent applications and multimodal services, this article refers to the visual cognitive mechanism of the human brain to improve multimedia representation and communication efficiency and studies the architecture of semantic communication systems, privacy protection mechanisms, and quality evaluation and guarantee mechanisms. The “semantic-network-semantic” architecture was established to integrate semantic computing, intelligent communication, and endogenous security to improve transmission efficiency, service quality, and security credibility. Through prior semantic knowledge, semantic representation, and encoding transmission methods, the quality of service of semantic communication is improved. To build an efficient 6G semantic communication paradigm and system, the communication mode should be changed, and new ideas for future mobile communication should be provided.

关键词： 6G mobile communication Solid modeling Privacy computer architecture Semantic communication Brain modeling Encoding Security Multimedia communication Protection

来源：评论

学校读者我要写书评

暂无评论

HiRes: Hierarchical Feature Optimization and Rescorer for Automatic ICD Coding

HiRes: Hierarchical Feature Optimization and Rescorer for Au...

引用

2025 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2025

作者： Liang, Zhenpeng Guan, Hongjiao Lu, Wenpeng Peng, Xueping Xu, Bing Yang, Muyun Key Laboratory of Computing Power Network and Information Security Ministry of Education Shandong Computer Science Center National Supercomputer Center in Jinan Qilu University of Technology Shandong Academy of Sciences Jinan China Shandong Provincial Key Laboratory of Computing Power Internet and Service Computing Shandong Fundamental Research Center for Computer Science Jinan China Australian Artificial Intelligence Institute University of Technology Sydney Sydney Australia Faculty of Computing Harbin Institute of Technology Harbin China

ISBN: (纸本)9798350368741

The International Classification of Diseases (ICD) coding assigns standardized codes to diseases. Automating this process enhances the efficiency and accuracy of clinical records processing. However, current methods struggle with noisy and lengthy clinical texts, making it difficult to ensure the reliability of feature extraction. Furthermore, they typically make separate binary predictions for each code, overlooking the dependencies between them. To address these issues, we propose a novel model called Hierarchical Feature Optimization and Rescorer (HiRes). We employ a cascaded convolution architecture to mitigate noise and enhance feature representation. Additionally, a masked autoencoder-based rescorer is introduced to capture ICD code interdependencies, refining initial predictions for improved accuracy. The model also considers the inconsistencies in code representations. Experiments on the MIMIC datasets demonstrate the effectiveness of our model. © 2025 IEEE.

关键词： automatic ICD coding cascaded convolutional network multi-label classification

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：