检索结果-内蒙古大学图书馆

arXiv 2025年

作者： Zhang, Zherui Wang, Changwei Xu, Rongtao Xu, Wenhao Xu, Shibiao Zhang, Yu Guo, Li Zhou, Jie School of Artificial Intelligence Beijing University of Posts and Telecommunications Beijing China The State Key Laboratory of Multimodal Artificial Intelligence Systems Institute of Automation Chinese Academy of Sciences Beijing China The Key Laboratory of Computing Power Network and Information Security Ministry of Education Shandong Computer Science Center Qilu University of Technology Jinan China Shandong Provincial Key Laboratory of Computing Power Internet and Service Computing Shandong Fundamental Research Center for Computer Science Jinan China Tongji University Shanghai China

Data-Free Knowledge Distillation (DFKD) enables the knowledge transfer from the given pre-trained teacher network to the target student model without access to the real training data. Existing DFKD methods focus primarily on improving image recognition performance on associated datasets, often neglecting the crucial aspect of the transferability of learned representations. In this paper, we propose Category-Aware Embedding Data-Free Knowledge Distillation (CAE-DFKD), which addresses at the embedding level the limitations of previous rely on image-level methods to improve model generalization but fail when directly applied to DFKD. The superiority and flexibility of CAE-DFKD are extensively evaluated, including: i.) Significant efficiency advantages resulting from altering the generator training paradigm;ii.) Competitive performance with existing DFKD state-of-the-art methods on image recognition tasks;iii.) Remarkable transferability of data-free learned representations demonstrated in downstream tasks. © 2025, CC BY.

关键词： Contrastive Learning

来源：评论

学校读者我要写书评

暂无评论

An Encoder-Decoder Network with Residual Convolution for Leukocyte Image Segmentation

An Encoder-Decoder Network with Residual Convolution for Leu...

引用

International Conference on information Science and Control Engineering (ICISCE)

作者： Yuanyuan Chen Shenghua Teng Zuoyong Li Fuquan Zhang Fujian Provincial Key Laboratory of Information Processing and Intelligent Control College of Computer and Control Engineering Minjiang University Fuzhou China College of Electronics and Information Engineering Shandong University of Science and Technology Qingdao China Fuzhou Digital Healthcare Industry Technology Innovation Center Fuzhou China

The appearance analysis and counting of peripheral blood leukocytes can assist the diagnosis of blood diseases such as leukemia. Therefore, it is necessary to automatically extract leukocytes from blood smear images. In this paper, we propose an end-to-end peripheral blood leukocyte image segmentation method based on fully convolutional neural network, which can not only encode multi-scale contextual information, but also capture clear boundary information through decoding. Specifically, the proposed method first utilizes the context-aware feature encoder with residual convolutions to extract a set of multi-scale features. Then, a refinement module of parallel dilated convolutions with multiple dilation rates for expanding the receptive field of the feature maps is introduced to aggregate contextual information. Finally, clear prediction results are reconstructed by the context-aware feature decoder and skip connections. We demonstrate the effectiveness of the proposed method on two publicly available datasets.

关键词： Deep learning Image segmentation information science Aggregates Feature extraction Decoding Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

Research on Question Answering System Based on Bi-LSTM and Self-attention Mechanism

Research on Question Answering System Based on Bi-LSTM and S...

引用

International Conference on Industrial Engineering and Applications (ICIEA)

作者： Hao Xiang Jinguang Gu School of Computer Science and Technology Wuhan University of Science and Technology Key Laboratory of Intelligent Information Processing and Real-time Industrial System in Hubei Province Wuhan China

ISBN: (数字)9781728167855

ISBN: (纸本)9781728167862

With the development of artificial intelligence technology, intelligent question answering has become a hot research direction in the field of natural language processing. This paper proposes a question answering method based on Bi-LSTM and self-attention mechanism model. This method uses Bi-LSTM to encode and align the question and answer respectively, then uses self-attention to obtain the relationship between keywords, and finally performs softmax through the fully connected layer to obtain the similarity between the question and answer. Finally, in the experiment, compared with the traditional attention model, the accuracy rate of this model was increased by 1.6%, and the recall rate was increased by 1.5%.

关键词： Semantics Machine learning Task analysis Knowledge discovery Feature extraction Standards Training

来源：评论

学校读者我要写书评

暂无评论

Disease transmission in dynamic social networks constructed by reinforcement learning-driven preventive game

引用

Chaos, Solitons & Fractals 2025年 199卷

作者： Yinuo Qian Dawei Zhao Chengyi Xia School of Control Science and Engineering Tiangong University Tianjin 300387 China 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 250014 China Shandong Provincial Key Laboratory of Computer Networks Shandong Fundamental Research Center for Computer Science Jinan 250014 China School of Artificial Intelligence Tiangong University Tianjin 300387 China

Reinforcement learning delves into the intrinsic behaviors of multi-agents, while complex networks reveal their interactive relationships. In real life, an individual can be considered as a sophisticated multi-agent, whose behavioral choices are influenced by a myriad of factors. When contagious diseases penetrate a population, the adoption of preventive measures represents a manifestation of behavioral selection. This paper introduces reinforcement learning into the multi-agent decision within the network, leveraging the Q-learning algorithm to train individual decision-making. At each time step, individuals select the next preventive strategy based on their utility and strategies, and these choices subsequently alter the network structure, thereby influencing the prevalence of the disease. In addition, the incorporation of additional government incentives prompts individuals to opt for advanced preventive strategies, effectively containing disease transmission within the population. The construction of this model not only showcases the co-evolutionary process between network structure and disease transmission but also validates its efficacy through theoretical analysis and applications in real-world networks, offering a novel perspective towards the epidemic control.

关键词： Disease transmission Dynamic social networks Preventive strategy game Q-learning

来源：评论

学校读者我要写书评

暂无评论

Design and implementation of CNN custom processor based on risc-v architecture 21

Design and implementation of CNN custom processor based on r...

引用

21st IEEE International Conference on High Performance Computing and Communications, 17th IEEE International Conference on Smart City and 5th IEEE International Conference on Data Science and Systems, HPCC/SmartCity/DSS 2019

作者： Li, Zhenhao Hu, Wei Chen, Shuang College of Computer Science and Technology Wuhan University of Science and Technology China Hubei Province Key Laboratory of Intelligent Information Processing Real-time Industrial System China

ISBN: (纸本)9781728120584

With the rapid development of CNN(convolutional neural networks), the traditional CPU platform can not make full use of the parallelism of CNN. We decide to adopt a new and popular processor architecture: the risc-v architecture for experimental design. In this paper, a new convolutional neural network processor is designed based on risc-v architecture. The processor can take advantage of the parallelism of CNN and is more flexible. This paper completely designed a CNN processor based on the risc-v architecture. The processor uses a classic five-stage pipeline structure, and implements instruction buffer memory and data buffer memory, and adds peripherals such as FLASH, SRAM, and SDRAM. And, this paper designed custom instructions. Given the convolution operation frequently occurring in CNN, vector store instruction, vector load instruction, vector addition instruction, and convolution operation instruction are designed to accelerate the execution of the convolution process. The design has passed the simulation experiment. It can not only complete the general instructions but also run the custom instructions. The final simulation test verified the correctness of the design. © 2019 IEEE.

关键词： Convolution

来源：评论

学校读者我要写书评

暂无评论

Controlled Secure Direct Communication Protocol via the Three-Qubit Partially Entangled Set of States

引用

computers, Materials & Continua 2019年第3期58卷 809-827页

作者： Gang Xu Ke Xiao Zongpeng Li Xin-Xin Niu Michael Ryan School of Information Science and Technology North China University of TechnologyBeijing100144China Information Security Center State Key Laboratory of Networking and Switching TechnologyBeijing University of Posts and TelecommunicationsBeijing100876China School of Computer Science Wuhan UniversityHubei430072China Guizhou Provincial Key Laboratory of Public Big Data GuiZhou UniversityGuiyangGuizhou550025China School of Computing Dublin City UniversityDublin 9Ireland.

In this paper,we first re-examine the previous protocol of controlled quantum secure direct communication of Zhang et al.’s scheme,which was found insecure under two kinds of attacks,fake entangled particles attack and disentanglement ***,by changing the party of the preparation of cluster states and using unitary operations,we present an improved protocol which can avoid these two kinds of ***,the protocol is proposed using the three-qubit partially entangled set of *** is more efficient by only using three particles rather than four or even more to transmit one bit secret *** our using state is much easier to prepare for multiqubit states and our protocol needs less measurement resource,it makes this protocol more convenient from an applied point of view.

关键词： Controlled secure direct communication cluster state three-qubit partially entangled state entanglement swapping.

来源：评论

学校读者我要写书评

暂无评论

Data Analysis and Planning Analysis Management System Based on Tensorflow Machine Learning Algorithms

Data Analysis and Planning Analysis Management System Based ...

引用

2019 International Conference on Artificial Intelligence Technologies and Applications, ICAITA 2019

作者： Hu, Shuang Huang, Li Bai, Feng Zhang, Jun School of Computer Science and Technology Wuhan University of Science and Technology China Intelligent Information Processing and Real-time Industrial System Hubei Key Laboratory Institute China

At present, machine learning is quite popular, and tensorflow framework is also very popular. This system takes the plan management system as the practice platform, realizes the data analysis function through the machine learning understanding and the key point. Most of the current market are planning management system, but there is no self-analysis ability, need to base on data, there are limitations. Firstly, the system collects training data such as weight, body fat rate and so on, then builds a deep learning neural network, and finally runs the model to realize plan analysis, so as to help users analyze the feasibility of the plan and provide suggestions for users, and the analysis results will be displayed through processing. © 2019 IOP Publishing Ltd. All rights reserved.

关键词： Learning algorithms

来源：评论

学校读者我要写书评

暂无评论

Ordinal Preferences Driven Reputation Measurement for Online Services with User Incentive

Ordinal Preferences Driven Reputation Measurement for Online...

引用

IEEE International Conference on Web Services (ICWS)

作者： Xiaodong Fu Li Liu Lijun Liu Yong Feng Kun Yue Yunnan Provincial Key Laboratory of Computer Technology Application Faculty of Information Engineering and Automation Kunming University of Science and Technology Kunming China School of Information Science and Engineering Yunnan University Kunming China

ISBN: (数字)9781728187860

ISBN: (纸本)9781728187877

A core source of raw information used as inputs to the reputation systems of online services is the feedback ratings provided by users. However, it is impossible that all users rate services with the same criteria and so ratings of different users are incommensurable. Meanwhile, users are not necessarily willing to provide honest feedbacks. Thus, aggregating dishonest cardinal ratings into reputation will potentially lead to unreliable and misleading reputation. In this paper, we propose a reputation model that aggregates ordinal user preferences rather than cardinal ratings for online services with user incentive. A distance metric is defined to measure the discrepancy between ordinal preferences. Then an optimal reputation model with the attributes of incentive compatible and individually rational is proposed. We design a B&B algorithm to solve the optimization problem so that a reputation vector that maximizes the total value of all users can be found efficiently. A comprehensive experimental study and performance analysis are conducted to evaluate the effectiveness and efficiency of the proposed method.

关键词： Measurement Web services Conferences Aggregates Performance analysis Optimization

来源：评论

学校读者我要写书评

暂无评论

Ultrasound vascular image segmentation based on full convolutional aggregation neural network 3

Ultrasound vascular image segmentation based on full convolu...

引用

3rd International Conference on computer Science and Artificial Intelligence, CSAI 2019 and its Workshop - The 11th International Conference on information and Multimedia technology, ICIMT 2019

作者： Li, Yan Liu, Jun Hubei Province Key Laboratory of Intelligent Information Processing and Real-time Industrial Department of Computer Science and Technology Wuhan University of Science and technology Wuhan430070 China

ISBN: (纸本)9781450376273

The concept of ultrasound therapy has been proposed long ago. However, most of the previous methods of ultrasound therapy destroy tissue through thermal effects and cause great damage to patients. In recent years, ultrasound cavitation therapy has caused extensive discussion and research due to its unique noninvasiveness. In order to achieve real-time access to the patient during treatment, the lesion area must be processed simultaneously and accurately. However, ultrasound images mostly have lower signal-to-noise ratio, contrast, and blurred edges. In order to solve the segmentation problem, this paper proposes a new segmentation model Du-net based on the full convolutional neural network. Under the premise of deepening the network depth to obtain more information, the Encoder- Decoder method and the layered aggregation mode are used to prevent the gradient explosion. Solving boundary segmentation problem by constructing joint network, In the case of a small data set, the data enhancement method using a suitable data set effectively increases the usable image features and achieves better results on the ultrasound blood vessel image. © 2019 Association for Computing Machinery.

关键词： Convolution

来源：评论

学校读者我要写书评

暂无评论

Research on point aggregation algorithm based on WEBGL 3

Research on point aggregation algorithm based on WEBGL

引用

3rd IEEE International Conference on Electronic information technology and computer Engineering, EITCE 2019

作者： Chen, Jinghan Chen, Jianxun Wuhan University of Science and Technology Hubei Province Key Laboratory of Intelligent Information Processing and Real-time Industrial System College of Computer Science and Technology Wuhan China

ISBN: (数字)9781728135847

ISBN: (纸本)9781728135847

WebGL has the advantages of online interaction and hardware accelerated rendering to serve web maps. When there are a large number of spatial points in the visible area of the map that need to be visually expressed, the method of extracting and abstracting the distribution rules and intensiveness of the local spatial points can be achieved through technical synthesis of these spatial points, so as to achieve an objective and comprehensive display space by using limited areas. This paper focuses on the related technologies and principles of WebGL-based point aggregation visualization. The advantages and disadvantages of the algorithm are analyzed according to the aggregation process of the algorithm, and the algorithm is optimized. © 2019 IEEE.

关键词： Visualization

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：