检索结果-内蒙古大学图书馆

39th Annual AAAI Conference on Artificial Intelligence, AAAI 2025

作者： Gu, Shengbo Qiu, Yu-Kun Tang, Yu-Ming Wu, Ancong Zheng, Wei-Shi School of Computer Science and Engineering Sun Yat-sen University China Peng Cheng Laboratory Shenzhen China Key Laboratory of Machine Intelligence and Advanced Computing Ministry of Education China

ISBN: (纸本)157735897X

The generation of a virtual digital avatar is a crucial research topic in the field of computer vision. Many existing works utilize Neural Radiance Fields (NeRF) to address this issue and have achieved impressive results. However, previous works assume the images of the training person are available and fixed while the appearances and poses of a subject could constantly change and increase in real-world scenarios. How to update the human avatar but also maintain the ability to render the old appearance of the person is a practical challenge. One trivial solution is to combine the existing virtual avatar models based on NeRF with continual learning methods. However, there are some critical issues in this approach: learning new appearances and poses can cause the model to forget past information, which in turn leads to a degradation in the rendering quality of past appearances, especially color bleeding issues, and incorrect human body poses. In this work, we propose a maintainable avatar (MaintaAvatar) based on neural radiance fields by continual learning, which resolves the issues by utilizing a Global-Local Joint Storage Module and a Pose Distillation Module. Overall, our model requires only limited data collection to quickly fine-tune the model while avoiding catastrophic forgetting, thus achieving a maintainable virtual avatar. The experimental results validate the effectiveness of our MaintaAvatar model. © 2025, Association for the Advancement of Artificial Intelligence (***). All rights reserved.

关键词： Rendering (computer graphics)

来源：评论

学校读者我要写书评

暂无评论

Estimating Multiple Socioeconomic Attributes via Home Location-A Case Study in China

引用

Journal of Social computing 2021年第1期2卷 71-88页

作者： Shichang Ding Xin Gao Yufan Dong Yiwei Tong Xiaoming Fu State Key Laboratory of Mathematical Engineering and Advanced Computing Zhengzhou 276800China Department of Sociology Tsinghua UniversityBeijing 100085China Institute of Computer Science University of GottingenGottingen 37077Germany Shanghai Hejin Information Technology Company Shanghai 200100China

Inferring people’s Socioeconomic Attributes(SEAs),including income,occupation,and education level,is an important problem for both social sciences and many networked applications like targeted advertising and personalized *** works mainly focus on estimating SEAs from peoples’cyberspace behaviors and relationships,such as the content of tweets or the social networks between online *** cyberspace data,alternative data sources about users’physical behavior,like their home location,may offer new *** specifically,in this paper,we study how to predict a person’s income level,family income level,occupation type,and education level from his/her home *** a case study,we collect people’s home locations and socioeconomic attributes through a survey involving 9 provinces and 85 cities in *** further enrich home location with the knowledge from real estate websites,government statistics websites,online map services,*** learn a shared representation from input features as well as attribute-specific representations for different SEAs,we propose H2SEA,a factorization machine-based multi-task learning method with attention *** experiment results show that:(1)Home location can clearly improve the estimation accuracy for all SEA prediction tasks(e.g.,80.2%improvement in terms of F1-score in estimating personal income level);(2)The proposed H2SEA model outperforms alternative models for SEA inference in terms of various evaluation metrics,such as Area Under Curve(AUC),F-measure,and specificity;(3)The performance of specific SEA prediction tasks(e.g.,personal income)can be further improved if H2SEA only focuses on cities or villages due to urban-rural gap in China;(4)Compared with online crawled housing price data,the area-level average income and Points Of Interest(POI)are more important features for SEA inferences in China.

关键词： personal income family income occupation education multi-task learning

来源：评论

学校读者我要写书评

暂无评论

RaCMC: Residual-Aware Compensation Network with Multi-Granularity Constraints for Fake News Detection

arXiv

引用

arXiv 2024年

作者： Yu, Xinquan Sheng, Ziqi Lu, Wei Luo, Xiangyang Zhou, Jiantao School of Computer Science and Engineering Ministry of Education Key Laboratory of Information Technology Guangdong Province Key Laboratory of Information Security Technology Sun Yat-sen University Guangzhou510006 China State Key Laboratory of Mathematical Engineering and Advanced Computing Zhengzhou450002 China State Key Laboratory of Internet of Things for Smart City The Department of Computer and Information Science Faculty of Science and Technology University of Macau 999078 China

Multimodal fake news detection aims to automatically identify real or fake news, thereby mitigating the adverse effects caused by such misinformation. Although prevailing approaches have demonstrated their effectiveness, challenges persist in cross-modal feature fusion and refinement for classification. To address this, we present a residual-aware compensation network with multi-granularity constraints (RaCMC) for fake news detection, that aims to sufficiently interact and fuse cross-modal features while amplifying the differences between real and fake news. First, a multiscale residual-aware compensation module is designed to interact and fuse features at different scales, and ensure both the consistency and exclusivity of feature interaction, thus acquiring high-quality features. Second, a multi-granularity constraints module is implemented to limit the distribution of both the news overall and the image-text pairs within the news, thus amplifying the differences between real and fake news at the news and feature levels. Finally, a dominant feature fusion reasoning module is developed to comprehensively evaluate news authenticity from the perspectives of both consistency and inconsistency. Experiments on three public datasets, including Weibo17, Politifact and GossipCop, reveal the superiority of the proposed method. Copyright © 2024, The Authors. All rights reserved.

关键词： Fake detection

来源：评论

学校读者我要写书评

暂无评论

Adversarial Example Attacks Against Intelligent Malware Detection: A Survey

Adversarial Example Attacks Against Intelligent Malware Dete...

引用

International Conference on Applied Machine Learning (ICAML)

作者： Xuyan Qi Yonghe Tang Huanwei Wang Tieming Liu Jing Jing State Key Laboratory of Mathematic Engineering and Advanced Computing Zhengzhou China

With the advancement of information technology and the increasing prevalence of the internet, the number of malware has experienced an exponential, and the security threat to users' property and privacy is becoming more serious. With the great success of artificial intelligence especially deep learning technology in fields such as computer vision and speech processing, deep learning are gradually introduced into the field of malware detection, to process large amounts of malware more efficiently. However, in recent years, researchers have discovered that the deep learning model itself also has security risks. While malware detection based on deep learning brings convenience and technical progress, it also introduces some new threats, such as adversarial examples. This paper investigates adversarial example attacks against deep learning-based malware detectors. First, We reviewed research on malware detection techniques. Then we analyze the works related to adversarial example attacks against intelligent malware detectors. Finally, we discuss the challenges and prospects in future research.

关键词： Deep learning Computer vision Technological innovation Privacy Perturbation methods Detectors Predictive models

来源：评论

学校读者我要写书评

暂无评论

An Instance-Level Motion-Aware Graph Model for Multi-target Multi-camera Tracking

An Instance-Level Motion-Aware Graph Model for Multi-target ...

引用

International Joint Conference on Neural Networks (IJCNN)

作者： Xiaotong Fan Huicheng Zheng School of Computer Science and Engineering Sun Yat-sen University China Guangdong Key Laboratory of Information Security Technology China Key Laboratory of Machine Intelligence and Advanced Computing Ministry of Education China

ISBN: (数字)9798350359312

ISBN: (纸本)9798350359329

Multi-target multi-camera tracking (MTMCT), focusing on inferring trajectory across multiple surveillance videos, is holding significant practical utility. While numerous studies aim to learn visual features robust to illumination variation, occlusions, and other issues, the spatio-temporal information within a multi-camera system is not explored sufficiently. Existing methods generally exploit coarse-grained spatio-temporal information by modeling the distribution of time intervals between cameras. However, they tend to neglect the specific motion state of individual objects, which may hinder accurate cross-camera association. In this paper, we introduce a novel motionaware graph (MAG) model designed to extract instance-level spatio-temporal information that aligns with visual features and seamlessly aggregate these two kinds of information within a unified graph framework for MTMCT. Specifically, we propose a motion encoder-decoder module that predicts spatio-temporal consistency scores between objects based on their instance-level motion states. These scores are then integrated with visual similarity scores to generate discriminative feature representations for data association via a graph attention mechanism. Experimental evaluations and ablation studies on the large-scale MTA dataset demonstrate the superiority of our proposed model.

关键词： Visualization Attention mechanisms Tracking Surveillance Neural networks Lighting Predictive models

来源：评论

学校读者我要写书评

暂无评论

ZombieCoin3.0: On the Looming of a Novel Botnet Fortified by Distributed Ledger Technology and Internet of Things 23

ZombieCoin3.0: On the Looming of a Novel Botnet Fortified by...

引用

23rd IEEE International Conference on High Performance computing and Communications, 7th IEEE International Conference on Data Science and Systems, 19th IEEE International Conference on Smart City and 7th IEEE International Conference on Dependability in Sensor, Cloud and Big Data Systems and Applications, HPCC-DSS-SmartCity-DependSys 2021

作者： Gao, Haoyu Li, Leixiao Lin, Hao Chang, Xiangyang Wan, Jianxiong Li, Jie Zhu, Fangyuan College of Data Science and Application Inner Mongolia University of Technology China Inner Mongolia Autonomous Reg. Eng. and Technol. Research Center of Big Data Based Software Service China State Key Laboratory of Mathematical Engineering and Advanced Computing Zhengzhou China

ISBN: (纸本)9781665494571

P2P Botnet is famous for the resilience against termination. However, its dependence on Neighbor List (NL) makes it susceptible to infiltration and poison, also leading to a dearth of adequate protection of Botmaster's identity. So as to overcome these weaknesses, the novel botnet infrastructure combined with blockchain emerges. But due to some flaws such like high latency and expensive transactions fees on blockchain, these new schemes can't satisfy the need of constructing hyper large-scale botnet in practice. We propose ZombieCoin3.0, a Botnet underlying overlay leveraging IoT devices to be maintainers and integrated with Distributed Ledger Technology (DLT), which is able to deploy bots fully on blockchain. And it is versatile for file transfer, data feedback and other command and control scenarios that many contemporaries are impotent to cope with. Our proposal eliminates the P2P botnets' dependence on NL and the bottlenecks that other blockchain combined botnets have. Experiments' results authenticated that the effectiveness of our proposal as well as low latency of communication and nearly zero overhead of cost other comparatives unable to attain are achieved. © 2021 IEEE.

关键词： Costs Distributed ledger Botnet Blockchains Proposals Internet of Things Low latency communication

来源：评论

学校读者我要写书评

暂无评论

Exploring light-cone distribution amplitudes from quantum computing

引用

Science China(Physics,Mechanics & Astronomy) 2023年第8期66卷 57-65页

作者： Tianyin Li Xingyu Guo Wai Kin Lai Xiaohui Liu Enke Wang Hongxi Xing Dan-Bo Zhang Shi-Liang Zhu Guangdong Provincial Key Laboratory of Nuclear Science Institute of Quantum MatterSouth China Normal UniversityGuangzhou 510006China Guangdong-Hong Kong Joint Laboratory of Quantum Matter Southern Nuclear Science Computing Center South China Normal University.Guangzhou 510006China Department of Physics and Astronomy University of CaliforniaLos Angeles 90095USA Center of Advanced Quantum Studies Department of PhysicsBeijing Normal UniversityBejing 100875China Center for High Energy Physics Peking UniversityBeijing 100871China Guangdong-Hong Kong Joint Laboratory of Quantum Matter Frontier Research Institute for PhysicsSouth China Normal UniversityGuangzhou 510006China Guangdong Provincial Key Laboratory of Quantum Engineering and Quantum Materials School of Physics and Telecommunication EngineeringSouth China Normal UniversityGuangzhou 510006China

Light-cone distribution amplitudes(LCDAs)are essential nonperturbative quantities for theoretical predictions of exclusive highenergy processes in quantum chromodynamics(QCD).We demonstrate the prospect of calculating LCDAs on a quantum computer by applying a recently proposed quantum algorithm,with staggered fermions,to the simulation of the LCDA in the(1+1)-dimensional Nambu-Jona-Lasinio(NJL)model on classical *** agreement between the result from the classical simulation of the quantum algorithm and that from exact diagonalization justifies the proposed quantum *** find that the resulting LCDA in the NJL model exhibits features shared with the LCDAs obtained from the QCD.

关键词： light-cone distribution amplitude quantum algorithm NJL model

来源：评论

学校读者我要写书评

暂无评论

Machine learning study on time-temperature-transformation diagram of carbon and low-alloy steel

引用

Journal of Iron and Steel Research International 2023年第5期30卷 1032-1041页

作者： Xiao-ya Huang Biao Zhang Qiang Tian Hong-hui Wu Bin Gan Zhong-nan Bi Wei-hua Xue Asad Ullah Hao Wang School of Materials Science and Engineering University of Science and Technology BeijingBeijing 100083China School of Materials Science and Engineering Beihang UniversityBeijing 100191China High Temperature Materials Research Division Central Iron and Steel Research InstituteBeijing 100081China Beijing Advanced Innovation Center for Materials Genome Engineering Collaborative Innovation Center of Steel TechnologyUniversity of Science and Technology BeijingBeijing 100083China Beijing Key Laboratory of Advanced High Temperature Materials Central Iron and Steel Research InstituteBeijing 100081China School of Materials Science and Engineering Liaoning Technical UniversityFuxin 123099LiaoningChina Department of Mathematical Sciences Karakoram International UniversityGilgit-Baltistan 15100Pakistan

Time-temperature-transformation(TTT)diagram plays a critical role in designing appropriate heat treatment process of steels by describing the relationship among holding time,temperature,and quantities of phase *** predictions for TTT diagrams of new steel rapidly and accurately is therefore of much practical importance,especially for costly and time-consuming experimental ***,TTT diagrams for carbon and low-alloy steels were predicted using machine learning *** commonly used machine learning(ML)algorithms,backpropagation artificial neural network(BP network),LibSVM,k-nearest neighbor,Bagging,and Random tree,were adopted to select appropriate models for the *** results illustrate that Bagging is the optimal model for the prediction of pearlite transformation and bainite transformation,and BP network is the optimal model for martensite ***,the ML framework composed of Bagging and BP network models was applied to predict the entire TTT ***,the ML models show superior performance on the prediction of testing samples than the commercial software JMatPro.

关键词： Time-temperature-transformation diagram Carbon steel Low-alloy steel Machine learning Prediction framework

来源：评论

学校读者我要写书评

暂无评论

MDFA: A Malware Detection Framework Based on Attention Mechanism with Bi-LSTM

MDFA: A Malware Detection Framework Based on Attention Mecha...

引用

International Conference on Applied Machine Learning (ICAML)

作者： Xuyan Qi Huanwei Wang Yonghe Tang Wei Liu Jiangtao Xie State Key Laboratory of Mathematic Engineering and Advanced Computing Zhengzhou China

As the existing malware intelligent detection methods have shortcomings and low accuracy of manual feature extraction and feature processing, a Malware Detection Framework with Attention mechanism based on Bi-directional Long Short-Term Memory (MDFA) was proposed. To training this method, We have collected 15, 893 malware and 10, 959 benign program samples and constructed a dataset. Then, the sample data was labeled and sanitized by determining the results based on VirusTotal analysis. Finally, the Bi-directional Long Short-Term Memory model injected into the attention mechanism was built and trained to complete the malware detection. Experimental results show that the accuracy of MDFA was improved by at least 2. 3%, and the recall rate was improved significantly compared with the existing methods, verifying the effectiveness of MDFA.

关键词： Training Computer viruses Bidirectional control Manuals Machine learning Feature extraction Transformers

来源：评论

学校读者我要写书评

暂无评论

Exploration and research of blended learning in the "principles of computer organization" 2

Exploration and research of blended learning in the "princip...

引用

2nd International Conference on Computer Science and Educational Informatization, CSEI 2020

作者： Hou, Yifan Wang, Wei Liu, Fudong Shan, Zheng Wang, Junchao Wang, Lixin State Key Laboratory of Mathematical Engineering and Advanced Computing Information Engineering University Zhengzhou China

ISBN: (纸本)9781728170077

The'Principles of Computer Organization' is the core basic course for the construction of computer hardware courses. In this paper, it analyzes the characteristics of the 'Principles of Computer Organization 'Aiming at the challenges faced by the traditional teaching. This paper makes a beneficial exploration on the teaching reform of the Principle of Computer Composition based on the blended learning mode from the aspects of teaching content, teaching method transformation, interactive communication, evaluation and assessment, etc. This paper proposes a framework for self-learning ability training based on a blended learning model, which is applied and formed in the course of principles. Teaching practice shows that the reform of teaching mode and teaching practice not only stimulates learning interest and motivation of students, but also improves their self-learning ability and practical ability. © 2020 IEEE.

关键词： Teaching

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：