检索结果-内蒙古大学图书馆

Learning to Generate Diverse data From a Temporal Perspective for data-Free Quantization

IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS FOR VIDEO TECHNOLOGY 2024年第10期34卷 9484-9498页

作者： Luo, Hui Zhang, Shuhai Zhuang, Zhuangwei Mai, Jiajie Tan, Mingkui Zhang, Jianlin Chinese Acad Sci Natl Key Lab Opt Field Manipulat Sci & Technol Chengdu 610209 Peoples R China Chinese Acad Sci Inst Opt & Elect Key Lab Opt Engn Chengdu 610209 Peoples R China Univ Chinese Acad Sci Sch Elect Elect & Commun Engn Beijing 100049 Peoples R China South China Univ Technol Sch Software Engn Guangzhou 510641 Peoples R China South China Univ Technol Minist Educ Key Lab Big Data & Intelligent Robot Guangzhou 510641 Peoples R China

Model quantization is a prevalent method to compress and accelerate neural networks. Most existing quantization methods usually require access to real data to improve the performance of quantized models, which is often infeasible in some scenarios with privacy and security concerns. Recently, data-free quantization has been widely studied to solve the challenge of not having access to real data by generating synthetic data, among which generator-based data-free quantization is an important type. Previous generator-based methods focus on improving the performance of quantized models by optimizing the spatial distribution of synthetic data, while ignoring the study of changes in synthetic data from a temporal perspective. In this work, we reveal that generator-based data-free quantization methods usually suffer from the issue that synthetic data show homogeneity in the mid-to-late stages of the generation process due to the stagnation of the generator update, which hinders further improvement of the performance of quantized models. To solve the above issue, we propose introducing the discrepancy between the full-precision and quantized models as new supervision information to update the generator. Specifically, we propose a simple yet effective adversarial Gaussian-margin loss, which promotes continuous updating of the generator by adding more supervision information to the generator when the discrepancy between the full-precision and quantized models is small, thereby generating heterogeneous synthetic data. Moreover, to mitigate the homogeneity of the synthetic data further, we augment the synthetic data with linear interpolation. Our proposed method can also promote the performance of other generator-based data-free quantization methods. Extensive experimental results show that our proposed method achieves superior performances for various settings on data-free quantization, especially in ultra-low-bit settings, such as 3-bit.

关键词： Quantization (signal) data models Synthetic data Generators Computational modeling Training Analytical models Model quantization data-free quantization generation process synthetic data linear interpolation

来源：评论

学校读者我要写书评

暂无评论

Learning Situation Risk Cognition and Measurement Based on data-Driven 17th

Learning Situation Risk Cognition and Measurement Based on D...

引用

17th International Conference on Computer Science and Education, ICCSE 2022

作者： Mi, Chunqiao Deng, Qingyou Zhao, Changhua Yin, Dong Liu, Yiwen School of Computer and Artificial Intelligence Huaihua University Hunan Huaihua418000 China Key Laboratory of Wuling-Mountain Health Big Data Intelligent Processing and Application in Hunan Province Universities Hunan Huaihua418000 China Key Laboratory of Intelligent Control Technology for Wuling-Mountain Ecological Agriculture in Hunan Province Hunan Huaihua418000 China

ISBN: (纸本)9789819924455

At present, online teaching has become more and more popular especially in the context of the current epidemic, and quality governance has become the internal needs of modern education development, while there is no simple and easy to use learning situation risk cognition method for specific online teaching class. In order to deal with this problem, in this study a data-driven method of learning situation risk cognition and measurement for online teaching is provided, which uses student initiative degree, concentration degree, duration degree and interaction degree to measure comprehensive learning effective degree and reflect student learning situation risk in an online class. Besides, normalized score earned by student in knowledge point test after online class is used to validate the calculation method designed, and the obtained results show that it is promising and easy to calculation. It provides a basis for decision making of students’ learning situation risk early warning and also provides a data-driven management method for the guarantee of online teaching quality, which has both academic and practical significance. © 2023, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.

关键词： Students

来源：评论

学校读者我要写书评

暂无评论

TSKE: A Dual-Branch Model for Knowledge Graph Embedding with Joint Textual and Structural Information 11th

TSKE: A Dual-Branch Model for Knowledge Graph Embedding wit...

引用

11th International Joint Conference on Rough Sets, IJCRS 2025

作者： Deng, Weibin Xiong, Yuanyuan Jiang, Ziqun Yu, Hong Chongqing Key Laboratory of Computational Intelligence Chongqing University of Posts and Telecommunications Chongqing China Smart Posts Engineering Technology Research Center Chongqing University of Posts and Telecommunications Chongqing China Key Laboratory of Cyberspace Big Data Intelligent Security Ministry of Education Chongqing China

ISBN: (纸本)9783031927430

Knowledge graphs (KGs) serve as critical structured representations of the real world, modeling entities and their complex relational patterns. While traditional knowledge graph embedding (KGE) methods excel at capturing structural regularities through geometric or tensor operations, they have limitations in effectively integrating structural and textual information. To address this limitation, we propose TSKE, a novel dual-branch architecture that effectively integrates both textual and structural information. The text branch employs language models to encode textual descriptions, generating context-aware semantic representations. Simultaneously, the structure branch introduces a multi-scale dilated attention mechanism that captures local neighborhood patterns and global graph dependencies through hierarchical receptive fields. The efficient channel attention further enhances feature selectivity by adaptively filtering noise and emphasizing feature relevance. TSKE incorporates an adaptive fusion module that dynamically adjusts the contribution weights of textual and structural features. Experimental results demonstrate TSKE’s superior performance in KGE. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.

关键词： Dilated attention Knowledge graph embedding Link prediction

来源：评论

学校读者我要写书评

暂无评论

State of Health Estimation of Lithium-ion Battery Based On Improved Transfer Learning

State of Health Estimation of Lithium-ion Battery Based On I...

引用

第35届中国控制与决策会议

作者： Penghua Li Kangheng Shan Key Laboratory of Intelligent Computing for Big Data College of Automation Chongqing University of Posts and Telecommunications

Many existing methods of forecasting the stateof-health(SOH) assume that training and testing data follow the same *** model based on dataset under one working condition may be ineffective for the dataset under another working condition due to the distribution *** order to meet this challenge,this paper proposes an improved method Mutual Information Domain-Adversial Neural Networks(DANN) based on domain adaptation,which improves the domain discriminator to better extract domain invariant *** addition,to avoid the loss of target information,the mutual information among target features,source features,and original target data is calculated to fix the features on the target site during the migration *** from the traditional methods,we only use 40% of the data sets for training,and the rest are used for prediction,so we can complete the prediction of more *** results show that this method can accurately and stably predict SOH.

关键词： transfer learning state-of-health lithium-ion battery neural network

来源：评论

学校读者我要写书评

暂无评论

LAGIM:A Label-Aware Graph Interaction Model for Joint Multiple Intent Detection and Slot Filling

LAGIM:A Label-Aware Graph Interaction Model for Joint Multip...

引用

第35届中国控制与决策会议

作者： Penghua Li Ziheng Huang Key Laboratory of Intelligent Computing for Big Data College of Automation Chongqing University of Posts and Telecommunications

Multi-intent spoken language understanding joint model can handle multiple intents in an utterance and is closer to complicated real-world scenarios,attracting increasing ***,existing research(1) usually focuses on identifying implicit correlations between utterances and one-hot encoding while ignoring intuitive and explicit original label characteristics;(2) only considers the token-level intent-slot interaction,which results in the limitation of the *** this paper,we propose a Label-Aware Graph Interaction Model(LAGIM),which captures the correlation between utterances and explicit labels' semantics to deliver enriched ***,a global graph interaction module is constructed to model the sentence-level interaction between intents and ***,we propose a novel framework to model the global interactive graph based on the injection of the original label semantics,which can fuse explicit original label features and provide global *** results show that our model outperforms existing approaches,achieving a relative improvement of 11.9% and 2.1%overall accuracy over the previous state-of-the-art model on the MixATIS and MixSnips datasets,respectively.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Course Homework Reform in Universities Based on Extended Reality 17th

Course Homework Reform in Universities Based on Extended Rea...

引用

17th International Conference on Computer Science and Education, ICCSE 2022

作者： Fu, Jinrong Liu, Yiwen Mi, Chunqiao Peng, Xiaoning Xiao, Jianhua School of Computer and Artificial Intelligence Huaihua University Hunan Huaihua418000 China Key Laboratory of Wuling-Mountain Health Big Data Intelligent Processing and Application in Hunan Province Universities Hunan Huaihua418000 China Key Laboratory of Intelligent Control Technology for Wuling-Mountain Ecological Agriculture in Hunan Province Hunan Huaihua418000 China

ISBN: (纸本)9789819924486

In view of the serious issues commonly existing with coursework in Chinese universities at present, such as its original function weakening or being suppressed, its form being too abstract and lack of elaborate design. In order to resolve these problems observed in usual homework, this paper proposes a reform scheme for cloud coursework based on Extended Reality and Bloom model. It mainly includes five key parts, such as interaction, scene, comprehensive evaluation, digital twin and data analysis. In addition, the reform’s effectiveness is illustrated through amassing and comparing the feedback of college students for teaching. © 2023, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.

关键词： Augmented reality

来源：评论

学校读者我要写书评

暂无评论

sSecure Net: A Hybrid CNN-LSTM-based Intrusion Detection System for Securing IoT Networks 25

sSecure Net: A Hybrid CNN-LSTM-based Intrusion Detection Sys...

引用

Proceedings of the 4th International Conference on Computer, Artificial Intelligence and Control Engineering

作者： Ovi Abul Hussain Zigang Chen Haihua Zhu Chongqing Key Laboratory of Cyberspace Security Monitoring and Governance & Key Laboratory of Cyberspace Big Data Intelligent Security Ministry of Education Chongqing University of Posts and Telecommunications Chongqing China

ISBN: (纸本)9798400712647

The widespread deployment of IoT devices has seen the emergence of a huge security threat, and IoT networks are the favorite targets for numerous cyberattacks. Due to the scarce resources in most IoT devices, the conventional intrusion detection systems (IDS) are unable to catch up with the enormous traffic in IoT networks. This paper suggests a new intrusion detection model based on CNN-LSTM, incorporating deep learning and statistical filtering to increase the intrusion detection in IoT networks. The model uses Convolutional Neural Networks (CNN) to capture the spatial relationships and Long Short-Term Memory (LSTM) networks to capture the temporal relationships in IoT network traffic. The model also uses statistical filtering techniques, including Median Filtering and Standard Deviation-based Filtering, to pre-process the traffic and eliminate the noise and the outliers, making the model's precision improved. The model is tested based on the Edge-IIoTset dataset, comprised of different diverse attack types, including Denial of Service (DoS), SQL injection, Ransomware, and Man-in-the-Middle (MITM). The model gives 94.99% precision, reflecting the model's precision in detecting different intrusions. This paper sets the efficacy in the use of the integration between the use of CNN-LSTM and statistical filtering to build an efficient, scalable, and precise intrusion detection model in IoT networks. The findings offer a real-time solution to increase IoT security, in particular, intrusion detection in IoT networks in real-time.

关键词： Anomaly Detection

来源：评论

学校读者我要写书评

暂无评论

Cross-Ray Neural Radiance Fields for Novel-view Synthesis from Unconstrained Image Collections

Cross-Ray Neural Radiance Fields for Novel-view Synthesis fr...

引用

International Conference on Computer Vision (ICCV)

作者： Yifan Yang Shuhai Zhang Zixiong Huang Yubing Zhang Mingkui Tan South China University of Technology Pazhou Lab Guangzhou Shiyuan Electronics Co. Ltd Key Laboratory of Big Data and Intelligent Robot Ministry of Education

Neural Radiance Fields (NeRF) is a revolutionary approach for rendering scenes by sampling a single ray per pixel and it has demonstrated impressive capabilities in novel-view synthesis from static scene images. However, in practice, we usually need to recover NeRF from unconstrained image collections, which poses two challenges: 1) the images often have dynamic changes in appearance because of different capturing time and camera settings; 2) the images may contain transient objects such as humans and cars, leading to occlusion and ghosting artifacts. Conventional approaches seek to address these challenges by locally utilizing a single ray to synthesize a color of a pixel. In contrast, humans typically perceive appearance and objects by globally utilizing information across multiple pixels. To mimic the perception process of humans, in this paper, we propose Cross-Ray NeRF (CR-NeRF) that leverages interactive information across multiple rays to synthesize occlusion-free novel views with the same appearances as the images. Specifically, to model varying appearances, we first propose to represent multiple rays with a novel cross-ray feature and then recover the appearance by fusing global statistics, i.e., feature covariance of the rays and the image appearance. Moreover, to avoid occlusion introduced by transient objects, we propose a transient objects handler and introduce a grid sampling strategy for masking out the transient objects. We theoretically find that leveraging correlation across multiple rays promotes capturing more global information. Moreover, extensive experimental results on large real-world datasets verify the effectiveness of CR-NeRF. The code and data can be found at https://***/YifYang993/***.

关键词：

来源：评论

学校读者我要写书评

暂无评论

CQUPT-FL: Cross-Domain Sharing and Security Awareness and Early Warning Platform of Health Science big data

CQUPT-FL: Cross-Domain Sharing and Security Awareness and Ea...

引用

Medical Artificial Intelligence (MedAI), IEEE International Conference on

作者： Yunpeng Xiao Qunqing Zhang Wanjing Zhao Xufeng Li Jinhua Peng Haonan Mo Haipeng Zhu Fei Tang Key Laboratory of Big Data Intelligent Computing Chongqing University of Posts and Telecommunications Chongqing China

Federated learning (FL) is a rapidly growing research area in machine learning, but it is problematic. It has been questioned whether or not existing FL libraries are practical in the area of medical privacy. To address these issues, we developed the CQUPT-FL system. The system focuses on resolving the conflict between data integrity and medical data privacy protection in cross-domain and cross-institution collaborative analysis. CQUPT-FL supports distributed computing and stand-alone simulation computing methods. To deal with the problems of heterogeneity, data domain diversity, and effective data scarcity, we adopted key technologies such as multi-party secure computing and holistic information representation and studied user identification, privacy protection, and heterogeneous user alignment to achieve sustainable Cross-domain and cross-platform data fusion of letters. The goal of introducing the CQUPT-FL system is to improve the level of data privacy protection and enhance the data privacy protection mechanism, solve the machine learning dilemma in the field of medical privacy, and provide a reliable solution for cross-domain collaborative analysis.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Few-Shot Object Detection Based On Label Constrained data Augmentation

Few-Shot Object Detection Based On Label Constrained Data Au...

引用

International Conference on big data and Information Analytics (bigDIA)

作者： Yan Shi Guoyin Wang Qun Liu Key Laboratory of Big Data Intelligent Computing Chongqing University of Posts and Telecommunications Chongqing China

The majority of object detection methods typically depend on a significant quantity of annotated data, while few-shot object detection (FSOD) endeavors to identify novel classes of objects using a limited number of training instances. However, the limited number of samples leads to the problem of disparate data distributions between the source and target domains, which makes the generalization ability of the detector usually weak. In this paper, we combine data augmentation with fine-tuning to design a pseudo-labeled constrained model called LCDA, aiming to obtain high-quality pseudo-labels to effectively enrich the training examples. Furthermore, we leverage the pre-trained CLIP model to enhance the quality of pseudo-labels by restricting the category information as well as the designed bounding box consistency criterion. The experimental outcomes demonstrate that our model outperforms the existing models on two public datasets across various shot scenarios. The average enhancement of our method on different shots is 1.2AP, 1.6AP, 2.3AP, 2.7AP, and 4.2AP. We also validate the performance of the model in real applications of the USV dataset which shows an improvement of 1.9AP over the baseline methods. All demonstrate the effectiveness of our model.

关键词：

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：