作者： Handong Ma¹;Changsheng Li²;Xinchu Shi³;Ye Yuan²;Guoren Wang²; （¹School of Computer Science and Engineering, University of Electronic Science and Technology of China, Chengdu, China ²School of Computer Science and Technology, Beijing Institute of Technology, Beijing, China ³Meituan Group, Beijing, China）

出处： IEEE Transactions on Neural Networks and Learning Systems 2024 Vol.35 No.2 P2894-2900

关键词： Learning systems;Decoding;Task analysis;Deep learning;Data models;Training;Feature extraction;Active Learning;Representative Sample;Deep Learning;Deep Models;Unsupervised Learning;Representation Learning;Graph Structure;Well-known Problem;Unsupervised Model;Shortcut Connection;Graph Learning;Nearest Neighbor Graph;Unsupervised Deep Learning;Unsupervised Learning Model;Data Structure;Matrix Multiplication;Latent Space;Regularization Term;Nodes In The Graph;Symmetric Structure;Latent Representation;Graph Neural Networks;Logistic Regression Classifier;Trade-off Parameter;Robust Representation;Reproducing Kernel Hilbert Space;Graph Laplacian;SVM Classifier

摘要： Recently, deep learning has been successfully applied to unsupervised active learning. However, the current method attempts to learn a nonlinear trans ...

作者： Wentian Zhao¹;Xinxiao Wu^1,2; （¹Beijing Key Laboratory of Intelligent Information Technology, School of Computer Science and Technology, Beijing Institute of Technology, Beijing, China ²Guangdong Laboratory of Machine Perception and Intelligent Computing, Shenzhen MSU-BIT University, Shenzhen, China）

出处： IEEE Transactions on Multimedia 2024 Vol.26 P2659-2670

关键词： Visualization;Internet;Knowledge graphs;Online services;Encyclopedias;Knowledge engineering;Knowledge based systems;Image Captioning;Multimodal Knowledge;Knowledge Base;Visual Cues;Extensive Experiments;Attention Mechanism;Image Object;Background Knowledge;Effective Description;Visual Object;External Knowledge;Graph Attention;Transformer;Results In Table;Object Detection;Intersection Over Union;Bounding Box;Directed Graph;News Articles;Human Faces;Event Image;Graph Attention Network;Object Detection Model;Google Images;Pre-trained Language Models;Sentence Level;Text Of Articles;Related Entities;Word Level;Names Of People

摘要： Entity-aware image captioning aims to describe named entities and events related to the image by utilizing the background knowledge in the associated ...

作者： Zongyao Hu¹;Lixiong Liu¹;CA1;Qingbing Sang²;Chongwen Wang³; （¹Beijing Laboratory of Intelligent Information Technology, School of Computer Science and Technology, Beijing Institute of Technology, Beijing 100081, China;²School of Artificial Intelligence and Computer Science, Jiangnan University, Wuxi 214122, China;³Beijing Key Laboratory of Digital Performance and Simulation Technology, School of Computer Science and Technology, Beijing Institute of Technology, Beijing 100081, China）

出处： Knowledge-Based Systems 2024 Vol.293 P111655

摘要： Most currently developed video quality assessment (VQA) algorithms have achieved excellent performance by using deep neural network (DNN). However, DN ...

作者： Linmei Hu¹;Zeyi Liu²;Ziwang Zhao²;Lei Hou³;Liqiang Nie⁴;Juanzi Li³; （¹School of Computer Science and Technology, Beijing Institute of Technology, Beijing, China ²School of Computer Science, Beijing University of Posts and Telecommunications, Beijing, China ³Department of Computer Science and Technology, Tsinghua University, Beijing, China ⁴School of Computer Science and Technology, Harbin Institute of Technology (Shenzhen), Shenzhen, Guangdong, China）

出处： IEEE Transactions on Knowledge and Data Engineering 2024 Vol.36 No.4 P1413-1430

关键词： Task analysis;Natural language processing;Training;Taxonomy;Linguistics;Computational modeling;Surveys;Language Model;Pre-trained Language Models;Natural Language;Text Data;Promising Direction;Language Understanding;Large Corpus;Self-supervised Learning;External Knowledge;Linguistic Knowledge;Natural Language Processing Tasks;Text Generation;Natural Language Understanding;Fine-tuning Stage;Model Performance;Structural Information;External Sources;Forms Of Knowledge;Long Short-term Memory;Pre-training Stage;Commonsense Knowledge;Input Text;Pre-training Tasks;Syntax Tree;Named Entity Recognition;Different Types Of Knowledge;Sources Of Heterogeneity;Source Of Knowledge;Explicit Reasoning

摘要： Pre-trained Language Models (PLMs) which are trained on large text corpus via self-supervised learning method, have yielded promising performance on v ...

作者： Zhendong Chen^1,2;1;Siu Cheung Hui³;2;Lejian Liao^1,2;CA3;Heyan Huang^1,2;4 （¹Beijing Engineering Research Center of High Volume Language Information Processing and Cloud Computing Applications, China;²School of Computer Science and Technology, Beijing Institute of Technology, China;³School of Computer Science and Engineering, Nanyang Technological University, Singapore）

出处： Neurocomputing 2024 Vol.583 P127549

摘要： As online rumors have the potential to greatly affect areas such as social order, stock prices, and presidential elections, there is an emerging neces ...

作者： Yuheng Shi¹;Xinxiao Wu^1,2;Hanxi Lin¹;Jiebo Luo³; （¹Beijing Laboratory of Intelligent Information Technology, School of Computer Science, Beijing Institute of Technology, Beijing, China ²Guangdong Provincial Laboratory of Machine Perception and Intelligent Computing, Shenzhen MSU-BIT University, Shenzhen, China ³Department of Computer Science, University of Rochester, Rochester, NY, USA）

出处： IEEE Transactions on Multimedia 2024 P1-11

关键词： Videos;Proposals;Semantics;Task analysis;Visualization;Training;Text recognition

摘要： Few-shot action recognition in videos is challenging as the lack of supervision makes it extremely difficult to generalize well to unseen actions. To ...

作者： Yuanzhang Li¹;Tianchi Sha¹;Thar Baker²;CA1;Xiao Yu³;CA2;Zhiwei Shi⁴;Sikang Hu¹; （¹School of Computer Science and Technology, Beijing Institute of Technology, 100081, Beijing, China;²Department of Computer Science, University of Sharjah, Sharjah, UAE;³Department of Computer Science and Technology, Shandong University of Technology, 255022, Zibo, China;⁴China Information Technology Security Evaluation Center, 100085, Beijing, China）

出处： Computing 2024 Vol.106 No.4 P1081-1097

关键词： Federated learning;Transfer learning;Convolutional neural network;Machine learning

摘要： To bring more intelligence to edge systems, Federated Learning (FL) is proposed to provide a privacy-preserving mechanism to train a globally shared m ...

作者： Tianhong Quan¹;Ye Yuan²;Yu Luo³;Youyi Song⁴;Teng Zhou⁵;Jiaqi Wang⁶; （¹School of Computer Science and Technology, Beijing Institute of Technology, Beijing, China ²College of Engineering, Shantou University, Shantou, China ³School of Computer Science and Technology, Guangdong University of Technology, Guangzhou, China ⁴School of Biomedical Engineering, Shenzhen University, Shenzhen, China ⁵School of Cyberspace Security (School of Cryptology), Hainan University, Haikou, China ⁶School of Pharmaceutical Sciences (Shenzhen), Sun Yat-sen University, Shenzhen, China）

出处： IEEE Transactions on Industrial Informatics 2024 Vol.20 No.3 P4137-4148

关键词： Task analysis;Compounds;Drugs;Machine learning;Inhibitors;Data augmentation;Training;Drug Screening;Machine Learning;Potent Inhibitor;Learning Algorithms;General Framework;Small Datasets;Real-world Applications;Data Augmentation;Exact Value;Real-world Datasets;Simple Classification;Order Statistics;Visfatin;Data Augmentation Methods;Sparse Coding;Adaptive Optimization;non-Gaussian Noise;Data-driven Machine Learning;Exact Prediction;Training Set;Inhibitory Potential Of Compounds;Classification Task;Real-world Data;Dataset Compounds;Second-order Statistics;Membership Function;Binary Label;Robust Classification;Label Noise;Unknown Samples

摘要： Data-driven machine learning is increasingly involved in human life and industrial development due to its large-scale testing and low time cost. Howev ...

作者： Huabing Liu¹;Dong Nie²;Jian Yang³;Jinda Wang⁴;Zhenyu Tang⁵; （¹School of Computer Science and Engineering, Beihang University, Beijing, China ²Department of Computer Science, University of North Carolina at Chapel Hill, Chapel Hill, NC, USA ³Beijing Engineering Research Center of Mixed Reality and Advanced Display, School of Optics and Electronics, Beijing Institute of Technology, Beijing, China ⁴Sixth Medical Center, Chinese PLA General Hospital, Beijing, China ⁵School of Computer Science and Engineering and the State Key Laboratory of Virtual Reality Technology and System, Beihang University, Beijing, China）

出处： IEEE Journal of Biomedical and Health Informatics 2024 Vol.28 No.3 P1484-1493

关键词： Image segmentation;Feature extraction;Gray-scale;Image registration;Biomedical imaging;Brain;Deep learning;Deep Learning;Feature Warping;Neuroimaging;Image Features;Image Segmentation;Brain Magnetic Resonance Imaging;Target Image;Regularization Term;Segmentation Results;Segmentation Accuracy;Segmentation Task;Differential Modulation;Public Image;Registration Method;Anatomical Knowledge;Medical Image Segmentation;Traditional Segmentation;Brain Parcellation;Final Segmentation Results;Smoothness Regularization;Backbone Segments;Segmentation Method;Brain Regions;Grayscale Images;Image Patches;Patch Size;Image Registration;Distributed Brain Regions;Neighboring Voxels;Image Intensity;Humans;Deep Learning;Brain;Image Processing, Computer-Assisted;Magnetic Resonance Imaging

摘要： Deep learning based multi-atlas segmentation (DL-MA) has achieved the state-of-the-art performance in many medical image segmentation tasks, e.g., bra ...

作者： Yuming Tang^1,#;, Yitian Zhang^2,#;, Tao Niu¹;, Zhen Li^2,3,;;, Zijian Zhang^1,3;, Huaping Chen⁴;, Long Zhang⁴; （¹School of Cyberspace Science & Technology, Beijing Institute of Technology, Beijing, 100081, China;²School of Computer Science & Technology, Beijing Institute of Technology, Beijing, 100081, China;³Southeast Institute of Information Technology, Beijing Institute of Technology, Putian, 351100, China;⁴Qianxin Technology Group Co., Ltd., Beijing, 100044, China）

出处： Computer Modeling in Engineering & Sciences 2024 Vol.139 No.3 P2451-2477

关键词： Federated learning;blockchain;privacy-preserving

摘要： Federated Learning (FL), as an emergent paradigm in privacy-preserving machine learning, has garnered significant interest from scholars and engineers ...