检索结果-内蒙古大学图书馆

International Conference on Intelligent Human-Machine Systems and Cybernetics, IHMSC

作者： Baixi Xing Xinjie Song Qimeng Chen Lei Shi Yanhong Pan Kaiqi Wang Mengyue Tang Industrial Design Institute Zhejiang University of Technology Hangzhou China School of Design and Architecture Zhejiang University of Technology Hangzhou China School of Computer Science and Technology Hangzhou Dianzi University Hangzhou China Newcastle University Newcastle upon Tyne UK

Users' viewing behavior could affect their perception and evaluation of design works. Taking into account users' visual attention as a subjective cognition cue, we used eye-tracking evidence to identify users' focus areas for further analysis. We conducted experiments to extract the image features of design images and the reviewers' eye-tracking data, aiming to predict the product design ranking in the competition through fusion data analysis. In particular, we collected 1,504 product design images from a design competition. Four deep convolutional neural networks were selected to explore the best aesthetics computation model. The experimental results show that using design images and eye-tracking data fusion can improve the model prediction performance. Finally, MobileNet-V3 achieves the highest classification accuracy of 74.75%. This suggests the proposed method can provide useful insights into personalized aesthetics evaluation and user-centered design perception.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Brain Cortical Surface Can Predict Fiber Trajectories

Brain Cortical Surface Can Predict Fiber Trajectories

引用

IEEE International Symposium on Biomedical Imaging

作者： Tianle Han Yifan Lv Mengshen He Yiheng Liu Tianming Liu Bao Ge School of Physics & Information Technology Shaanxi Normal University Xi'an China Department of Computer Science and Bioimaging Research Center Cortical Architecture Imaging and Discovery Lab The University of Georgia GA USA

ISBN: (数字)9798350313338

ISBN: (纸本)9798350313345

Gray matter and white matter are fundamental constituents of the brain, and exploring the relationship between them is for gaining insights into the intricate architecture of the brain. This study utilizes information extracted from the gray matter surface of the brain to predict the trajectories of white matter fibers. Specifically, it predicts the coordinates of fiber trajectories by utilizing both fiber endpoints and reference points on the cortical surface. Both qualitative analysis and extensive quantitative experiments confirm that information derived from the brain's cortical surface can be employed to predict fiber trajectories. Furthermore, this conclusion remains robust across diverse fiber tracking and surface reconstruction techniques. The study introduces a novel perspective, elucidating the fundamental principles of fiber connectivity and the intricate relationship between gray matter and white matter.

关键词： Surface reconstruction Brain Pathology Nerve fibers Magnetic resonance imaging Grey matter Streaming media

来源：评论

学校读者我要写书评

暂无评论

Energy-Efficient and Reliable Federated Learning in Heterogeneous Mobile-Edge Computing

Energy-Efficient and Reliable Federated Learning in Heteroge...

引用

Symposia and Workshops on Ubiquitous, Autonomic and Trusted Computing, UIC-ATC

作者： Yan Xu Liying Li Peijin Cong Junlong Zhou School of Computer Science and Engineering Nanjing University of Science and Technology China State Key Laboratory of Computer Architecture Institute of Computing Technology Chinese Academy of Sciences China

As a decentralized machine learning paradigm, Federated Learning (FL) is an emerging technique to protect user data privacy in Mobile Edge Computing (MEC). FL adopts the idea of distributed privacy computing, enabling the terminal devices to train machine learning models required by servers locally and upload model parameters to servers for aggregation. However, in the process of coupling FL with MEC systems, device heterogeneity, malicious use of poisoned data by users and limited system resources will seriously affect FL training accuracy, system reliability and energy efficiency. In this paper, a highly reliable and energy-efficient FL scheme is designed to solve these issues. Specifically, we first propose a calculation method for estimating the reliability of heterogeneous devices. Then, we design a highly reliable device selection scheme based on devices’ reliability, computing power, and participation times. Finally, we utilize the slack time in FL training to dynamically adjust the voltage and frequency of devices for reducing the energy consumption of training. Experimental results show that our proposed scheme improves the accuracy by up to 36.45% on average and saves energy by up to 48.63% when compared to the baseline and benchmarking methods.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Haze Risk Assessment and Trend Evolution in Fenwei Plain Based on PCA-XGBoost

Haze Risk Assessment and Trend Evolution in Fenwei Plain Bas...

引用

Environmental science and Green Energy (ICESGE), International Conference on

作者： Hongbin Dai Guangqiu Huang Huibin Zeng Rongchuan Yu Jingjing Wang School of Management Xi'an University of Architecture and Technology Xi'an China School of Mathematics and Computer Science Guangxi Science & Technology Normal University Laibin China

According to the regional disaster system theory, haze risk assessment indicators for the Fenwei Plain were chosen to effectively study the urban haze hazards in terms of the sensitivity of the disaster-causing environment, the hazard of haze components, and the vulnerability of disaster-bearing bodies. The risk level of haze hazards in Fenwei Plain cities was calculated using Principal Component Analysis (PCA). A quantitative assessment of the haze risk in the Fenwei Plain urban agglomeration was carried out using the extreme gradient boosting method (XGBoost) to investigate the spatial and temporal distribution patterns of PM2.5 and its trends. The results show that Xi'an exhibited the highest haze risk index and the PCA-XGBoost model performed best in the three evaluation indicators. The main factors affecting haze are the economy and population density.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Distilling Object Detectors with Feature Richness

arXiv

引用

arXiv 2021年

作者： Du, Zhixing Zhang, Rui Chang, Ming Zhang, Xishan Liu, Shaoli Chen, Tianshi Chen, Yunji University of Science and Technology of China China SKL of Computer Architecture Institute of Computing Technology CAS Beijing China Cambricon Technologies China University of Chinese Academy of Sciences China

In recent years, large-scale deep models have achieved great success, but the huge computational complexity and massive storage requirements make it a great challenge to deploy them in resource-limited devices. As a model compression and acceleration method, knowledge distillation effectively improves the performance of small models by transferring the dark knowledge from the teacher detector. However, most of the existing distillation-based detection methods mainly imitating features near bounding boxes, which suffer from two limitations. First, they ignore the beneficial features outside the bounding boxes. Second, these methods imitate some features which are mistakenly regarded as the background by the teacher detector. To address the above issues, we propose a novel Feature-Richness Score (FRS) method to choose important features that improve generalized detectability during distilling. The proposed method effectively retrieves the important features outside the bounding boxes and removes the detrimental features within the bounding boxes. Extensive experiments show that our methods achieve excellent performance on both anchor-based and anchor-free detectors. For example, RetinaNet with ResNet-50 achieves 39.7% in mAP on the COCO2017 dataset, which even surpasses the ResNet-101 based teacher detector 38.9% by 0.8%.Our implementation is available at https://***/duzhixing/FRS. © 2021, CC BY.

关键词： Object detection

来源：评论

学校读者我要写书评

暂无评论

Radio spectrum awareness using deep learning: Identification of fading channels, signal distortions, medium access control protocols, and cellular systems

引用

Intelligent and Converged Networks 2021年第1期2卷 16-29页

作者： Yu Zhou Hatim Alhazmi Mohsen H.Alhazmi Alhussain Almarhabi Mofadal Alymani Mingju He Shengliang Peng Abdullah Samarkandi Zikang Sheng Huaxia Wang Yu-Dong Yao Department of Electrical and Computer Engineering Stevens Institute of TechnologyHobokenNJ 07030USA the College of Information Science and Engineering Huaqiao UniversityXiamen 361021China the College of Engineering Architecture and TechnologyOklahoma State UniversityStillwaterOK 74078-1010USA

Radio spectrum awareness,including understanding radio signal activities,is crucial for improving spectrum utilization,detecting security vulnerabilities,and supporting adaptive *** tasks include spectrum sensing,identifying systems and terminals,and understanding various protocol *** this paper,we investigate various identification and classification tasks related to fading channel parameters,signal distortions,Medium Access Control(MAC)protocols,radio signal types,and cellular ***,we utilize deep learning methods in those identification and classification *** evaluations demonstrate the effectiveness of deep learning in those radio spectrum awareness tasks.

关键词： cellular system deep learning signal classification spectrum awareness Convolutional Neural Network(CNN)

来源：评论

学校读者我要写书评

暂无评论

Association of PM_(2.5)Reduction with Improved Kidney Function:A Nationwide Quasiexperiment among Chinese Adults

引用

Health Data science 2022年第1期2022卷 115-123页

作者： Yiqun Han Tao Xue Frank J.Kelly Yixuan Zheng Yao Yao Jiajianghui Li Jiwei Li Chun Fan Pengfei Li Tong Zhu Environmental Research Group MRC Centre for Environment and HealthImperial College LondonLondonUK BIC-ESAT and SKL-ESPC College of Environmental Sciences and EngineeringCenter for Environment and HealthPeking UniversityBeijing 100871China Institute of Reproductive and Child Health/Ministry of Health Key Laboratory of Reproductive Health and Department of Epidemiology and Biostatistics School of Public HealthPeking UniversityBeijingChina Center of Air Quality Simulation and System Analysis Chinese Academy of Environmental Planning100012 BeijingChina China Center for Health Development Studies Peking UniversityBeijingChina College of Computer Science and Technology Zhejiang UniversityHangzhouChina Computer Center Peking University and Peng Cheng LaboratoryChina Advanced Institute of Information Technology Peking UniversityChina

*** evidence from human studies has revealed the adverse impact of ambient fine particles(PM_(2.5))on health outcomes related to metabolic disorders and distant *** exposure to ambient PM_(2.5)leads to kidney impairment remains *** rapid air quality improvement driven by the clean air actions in china since 2013 provides an opportunity for a quasiexperiment to investigate the beneficial effect of PM_(2.5)reduction on kidney *** on two repeated nationwide surveys of the same population of 5115 adults in 2011 and 2015,we conducted a difference-indifference *** in long-term exposure to ambient PM_(2.5)were associated with changes in kidney function biomarkers,including estimated glomerular filtration rate by serum creatinine(GFRscr)or cystatin C(GFRcys),blood urea nitrogen(BUN),and uric acid(UA).*** a 10μg/m^(3)reduction in PM_(2.5),a significant improvement was observed for multiple kidney functional biomarkers,including GFRscr,BUN and UA,with a change of 0.42(95%confidence interval[CI]:0.06,0.78)mL/min/1.73m^(2),-0.38(-0.64,-0.12)mg/dL,and-0.06(-0.12,-0.00)mg/dL,respectively.A lower socioeconomic status,indicated by rural residence or low educational level,enhanced the adverse effect of PM_(2.5)on kidney *** results support a significant nephrotoxicity of PM_(2.5)based on multiple serum biomarkers and indicate a beneficial effect of improved air quality on kidney function.

关键词： Kidney organs ambient

来源：评论

学校读者我要写书评

暂无评论

Linfa-Q: Accurate Q-Learning with Linear Function Approximation

SSRN

引用

SSRN 2023年

作者： Wang, Zhechao Fu, Qiming Chen, Jianping Liu, Quan Wu, Hongjie Lu, You Hu, Fuyuan School of Electronics and Information Engineering Suzhou University of Science and Technology Suzhou215009 China Jiangsu Province Key Laboratory of Intelligent Building Energy Efficiency Suzhou University of Science and Technology Suzhou215009 China School of Architecture and Urban Planning Suzhou University of Science and Technology Suzhou215009 China School of Computer Science and Technology Soochow University Jiangsu Suzhou215006 China

Although Q-learning has achieved remarkable success in some practical cases, it often suffers from the overestimation problem in stochastic environments, which is commonly viewed as a shortcoming for Q-learning. Overestimated values are introduced by estimations on the next state Q value, which is well-known as the maximization bias. In this paper, we propose a more accurate method of the Q value estimation by the Q value decomposition and re-evaluation with similar samples based on linear function approximation. Specifically, we reform the parameterized incremental update formula of Q-learning and also prove that the new formula is equivalent to the original one. Moreover, we propose a new parameterized incremental update formula of Q-learning to deal with the overestimation problem and present the more accurate computing method, which can be used in those problems with continuous state space and stochastic environments. Experimentally, compared with other Q-learning based algorithms, the new algorithm can update the Q value faster and more accurately in the long term. Furthermore, the algorithm is robust to the learning rate, which can help users reduce adjustment times of the hyper-parameter. © 2023, The Authors. All rights reserved.

关键词： Reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

Disentanglement of Human Facial Features and the Inspection of Feature Extractor's Quality Based on TL-GAN

Disentanglement of Human Facial Features and the Inspection ...

引用

2021 International Conference on Applied Mathematics, Modelling and Intelligent Computing, CAMMIC 2021

作者： Gao, Yutong He, Shiwei Wang, Guanghan Xu, Peiyu Information and Computer Science Beijing University of Civil Engineering and Architecture Beijing102612 China Safety Engineering China University of Mining and Technology-Beijing Beijing100089 China Computer Science and Technology Tsinghua University Beijing100084 China Telecommunications Engineering and Management Beijing University of Posts and Telecommunications Hebei065000 China

Recent research on the Generated Adversarial Network models has made great progress in the task of generating images on human demand. Among them, Transparent Latent-space GAN tries to solve the problem by analyzing the latent space and finding the feature axes inside it. This model views the problem from a novel perspective, and it is easy to be carried out. However, more thorough and detailed experiments have shown that the original method on axes disentanglement of the model has certain drawbacks, which may lead to ethical problems, meaning that the model may pick up some pre-existing prejudice of racial or gender discrimination. This paper puts forward a new training mode aiming at reducing the bias existing in the generated images and simultaneously managing the disentanglement. Also, a more thorough examination of the model was conducted and it is found that the architecture of the model may be generalized to some universal functions. Among them, the quality evaluation of the feature extractor is the most practical and useful one. This finding, in turn, may be helpful for distinguishing a more accurate, effective, and robust feature extractor, thus improving the performance of the TL-GAN model in the first place. © Published under licence by IOP Publishing Ltd.

关键词：

来源：评论

学校读者我要写书评

暂无评论

A practical approach based on Bluetooth Low Energy and Neural Networks for indoor localization and targeted devices’ identification by smartphones

引用

Entertainment Computing 2022年第0期43卷

作者： Pau, Giovanni Arena, Fabio Collotta, Mario Kong, Xiangjie Kore University of Enna Faculty of Engineering and Architecture Enna94100 Italy Zhejiang University of Technology College of Computer Science and Technology Hangzhou310023 China

With the constant development of innovative technologies and the resulting growth of new services available on the market, applications that aim to control devices employing smartphones include more extensive selection menus increasingly. This condition can lead to non-intuitive use of the system. As a result, the user may reap an adverse experience while practicing these applications. An indoor tracking system must create a control method with a dynamic user interface and estimate the device pointed to by a smartphone. The use of Bluetooth Low Energy (BLE) and the fingerprinting technique, based on Received Signal Strength Indication (RSSI) values, can be a feasible solution for indoor localization. Furthermore, to obtain ever more precise indoor localization systems, there is continuous research on artificial neural networks’ employment as they adapt more quickly to changes in the RSSI values. This paper proposes a new approach to control devices through smartphones based on the joint application of BLE, fingerprinting, and neural networks. The system offers a dynamic user interface that changes according to the target device through sensors commonly located in modern smartphones. © 2022 Elsevier B.V.

关键词： Neural networks

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：