检索结果-内蒙古大学图书馆

24th IEEE international conference on High Performance computing and Communications, 8th IEEE international conference on Data Science and Systems, 20th IEEE international conference on Smart City and 8th IEEE international conference on Dependability in Sensor, Cloud and Big Data Systems and Application, HPCC/DSS/SmartCity/DependSys 2022

作者： Zuo, Zhiwei Cai, Qianqian Ji, Jungang Tang, Zhongan College of Computer Science and Electronic Engineering National Supercomputering Center in Changsha Hunan University Changsha China Szzt Electronics Co. Ltd Shenzhen China Third Surveying and Mapping Institute of Hunan Province Changsha China

ISBN: (纸本)9798350319934

Robustness is consistently the most crucial point for High Performance computing(HPC), machine learning, or large-scale machine learning facilitated by HPC. It is important to ensure a neural network is robust before exploring it in a safety-critical domain. Throughout the literature, we find that for non-differentiable activation functions such as ReLU, the correctness of some robust analysis approaches is still not guaranteed. To address this issue, our method relies on Clark's generalized gradients to derive local Lipschitz constants and interval arithmetic to effectively numerically validate existing methods for which the correctness of the results cannot be guaranteed. On this basis, extensive experiments are conducted to compare the Lipschitz constant results produced by our interval algorithm with previous methods for different parameter settings. We also perform convergence and error analysis on the experimental findings. The work done in this paper provides a reasonable basis for subsequent research on optimization algorithms. © 2022 IEEE.

关键词： machine learning

来源：评论

学校读者我要写书评

暂无评论

Predicting Cardiovascular Disease Through IoT and Deep learning Methods 4

Predicting Cardiovascular Disease Through IoT and Deep Learn...

引用

4th international conference on Sustainable Expert Systems, ICSES 2024

作者： Shukla, Sakshi Neduncheliyan, S. Bharath Institute of Higher Education and Research Chennai India School of Computing Bharath Institute of Higher Education and Research Chennai India

ISBN: (纸本)9798331540364

Today, heart disease affects many people and is a leading cause of death worldwide. Cardiovascular disease, also known as CVD, is the leading cause of morbidity and mortality, accounting for more than 70% of all deaths. According to the 2017 Global Burden of Disease Study, CVD causes 43% of all deaths. According to the World Health Organization (WHO), heart disease and stroke will kill 23.6 million people worldwide by 2030. This situation is difficult to predict when there are no symptoms and requires prior knowledge of the patient. A brilliant method called the Internet of Things captures the information that patients need and stores it in cloud servers. Many studies have been done on heart diseases;however, the accuracy of the prediction is very low. To solve this problem, we have created a revolutionary IoT-smart solution. Current healthcare systems are based on IoT devices and distributed systems, and they face serious problems due to insufficient distribution of patient information, which can lead to wrong diagnosis and treatment decisions. This paper uses a combination of ResNet197 and Gradient Boosting machine (GBM). ResNet197 is a deep residual neural network with 197 layers for CVD diagnosis. This technology outperforms existing models for predicting heart disease with 99.25% accuracy, 96.68% accuracy, 93.97 sensitivity, 96.96 F1 score, 98.69% accuracy, and 97.8% return. © 2024 IEEE.

关键词： Patient treatment

来源：评论

学校读者我要写书评

暂无评论

Beacon-based Indoor Fire Evacuation System using Augmented Reality and machine learning 6

Beacon-based Indoor Fire Evacuation System using Augmented R...

引用

6th IEEE international conference on Robotic computing, (IRC)

作者： Lee, Hwawon Chung, Dohyun Kim, Seongmin Lim, Jiwon Bahng, Yoonha Park, Suhyun Smith, Anthony H. Soongsil Univ Sch Software Seoul South Korea Chung Ang Univ Ind Secur Seoul South Korea Kangwon Natl Univ Comp Sci Chunchon South Korea Kwangwoon Univ Sch Software Seoul South Korea Chung Ang Univ Sch Comp Sci & Engn Seoul South Korea Paichai Univ Comp Engn Seoul South Korea Purdue Univ Comp & Informat Technol W Lafayette IN 47907 USA

ISBN: (纸本)9781665472609

The inefficiency of fire evacuation has been the issue since the present evacuation method is unsuitable for complex buildings. In order to improve the evacuation system, this paper aims at three main components. First, Kalman Filter and deep learning models were utilized to estimate the user's location accurately. Second, Q-learning based evacuation algorithm was designed to deal with various fire situations. Lastly, AR and a 2D map offer effective navigation systems. The proposed system offers the safest path based on accurate location with a user-friendly visual supplement.

关键词： Bluetooth Low Energy Deep learning Augmented Reality iBeacon Indoor Localization Real-time communication Reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

Improving Multilayer-Perceptron(MLP)-based Network Anomaly Detection with Birch Clustering on CICIDS-2017 Dataset 26

Improving Multilayer-Perceptron(MLP)-based Network Anomaly D...

引用

26th international conference on Computer Supported Cooperative Work in Design, CSCWD 2023

作者： Yin, Yuhua Jang-Jaccard, Julian Sabrina, Fariza Kwak, Jin Massey University Cybersecurity Lab School of Mathematical and Computational Sciences New Zealand Central Queensland University School of Engineering and Technology Australia Ajou University Department of Cyber Security Korea Republic of

ISBN: (纸本)9798350331684

The network intrusion threats are increasingly severe with the application of computer supported coorperative work. machine learning algorithms have been widely used in intrusion detection systems, including Multi-layer Perceptron (MLP). In this study, we proposed a two-stage model that combines the Birch clustering algorithm and MLP classifier to improve the performance of network anomaly multi-classification. In our proposed method, we first apply Birch or K-means as an unsupervised clustering algorithm to the CICIDS-2017 dataset to pre-group the data. The generated pseudo-label is then added as an additional feature to the training of the MLP-based classifier. The experimental results show that using Birch and K-Means clustering for data pre-grouping can improve intrusion detection system performance. Our method can achieve 99.73% accuracy in multi-classification using Birch clustering, which is better than similar researches using a stand-alone MLP model. © 2023 IEEE.

关键词： Intrusion detection

来源：评论

学校读者我要写书评

暂无评论

G-SEPM: Building an Accurate and Efficient soft Error Prediction Model for GPGPUs 21

G-SEPM: Building an Accurate and Efficient Soft Error Predic...

引用

international conference for High Performance computing, Networking, Storage and Analysis (SC21)

作者： Yue, Hengshan Wei, Xiaohui Li, Guangli Zhao, Jianpeng Jiang, Nan Tan, Jingweijia Jilin Univ Coll Comp Sci & Technol Changchun Jilin Peoples R China Univ Chinese Acad Sci SKL Comp Architecture ICT CAS Beijing Peoples R China

ISBN: (纸本)9781450384421

As GPUs become ubiquitous in large-scale general purpose HPC systems (GPGPUs), ensuring the reliable execution of such systems in the presence of soft errors is increasingly essential. To provide insights into how resilient GPU programs are toward soft errors, researchers typically rely on random Fault Injection (FI) to evaluate the tolerance of programs. However, it is expensive to obtain a statistically significant resilience profile and not suitable to identify all the error-critical fault sites of GPU programs. To address the above challenges, in this work, we build a GPGPU based soft Error Prediction Model (G-SEPM) that can replace FI to estimate the resilience characteristics of individual fault sites accurately and efficiently. We observe that the instruction-type, bit position, bit-flip direction, and error propagation information have capabilities to characterize fault site resiliency. Leveraging these heuristic features, G-SEPM drives the machine learning model to reveal the hidden interactions among fault site resiliency and our observed features. Experimental results demonstrate that G-SEPM achieves high accuracy for fault site error estimation and critical fault site identification while introducing negligible overhead. In addition, G-SEPM can provide essential insight for programmers/architects to design more cost-effective soft error mitigation solutions.

关键词： machine learning GPGPU fault injection error resilience soft error

来源：评论

学校读者我要写书评

暂无评论

Eco-FL: Adaptive Federated learning with Efficient Edge Collaborative Pipeline Training 51

Eco-FL: Adaptive Federated Learning with Efficient Edge Coll...

引用

51st international conference on Parallel Processing (ICPP)

作者： Ye, Shengyuan Zeng, Liekang Wu, Qiong Luo, Ke Fang, Qingze Chen, Xu Sun Yat Sen Univ Sch Comp Sci & Engn Guangzhou Peoples R China

ISBN: (纸本)9781450397339

Federated learning (FL) has been a promising paradigm in distributed machine learning that enables in-situ model training and global model aggregation. While it can well preserve private data for end users, to apply it efficiently on IoT devices yet suffer from their inherent variants: their available computing resources are typically constrained, heterogeneous, and changing dynamically. Existing works deploy FL on IoT devices by pruning a sparse model or adopting a tiny counterpart, which alleviates the workload but may have negative impacts on model accuracy. To address these issues, we propose Eco-FL, a novel Edge Collaborative pipeline based Federated learning framework. On the client side, each IoT device collaborates with trusted available devices in proximity to perform pipeline training, enabling local training acceleration with efficient augmented resource orchestration. On the server side, Eco-FL adopts a novel grouping-based hierarchical architecture that combines synchronous intra-group aggregation and asynchronous inter-group aggregation, where a heterogeneity-aware dynamic grouping strategy that jointly considers response latency and data distribution is developed. To tackle the resource fluctuation during the runtime, Eco-FL further applies an adaptive scheduling policy to judiciously adjust workload allocation and client grouping at different levels. Extensive experimental results using both prototype and simulation show that, compared to state-of-the-art methods, Eco-FL can upgrade the training accuracy by up to 26.3%, reduce the local training time by up to 61.5%, and improve the local training throughput by up to 2.6x.

关键词： Federated learning edge intelligence pipeline parallelism parallel processing

来源：评论

学校读者我要写书评

暂无评论

Classification of dermoscopic images using soft computing techniques

引用

NEURAL computing & APPLICATIONS 2021年第19期33卷 13015-13026页

作者： Maniraj, S. P. Sardarmaran, P. Sathyabama Inst Sci & Technol Sch Comp Dept Comp Sci & Engn Chennai 600119 Tamil Nadu India

Medical diagnosis using machine learning techniques has great attention over the last two decades. The detection of skin cancer based on visual information requires highly skilled dermatologists, and also it is a time-consuming process. To analyse the condition of patients and diagnose the diseases at the earliest, an automated classification system is needed that may help to enhance the clinical decision. There are many clinical trials available to classifying melanoma skin cancer. In this study, soft computing techniques are enabled such as image processing, Genetic Algorithm (GA) and Deep learning Neural Network (DLNN) to get accurate result of classification. To achieve this, three major modules are developed. The first part is dermoscopic image preprocessing from which the dermoscopic images can be prepared by removing noises and skin hairs. The second one is feature extraction and selection. The former one utilizes 3-dimensional Discrete Wavelet Transform (3DWT), and later one uses GA. The last module is the knowledge discover step where the dermoscopic images are classified using an appropriate DLNN classifier. In the result of comparative analysis, a maximum accuracy of 98.67% is obtained using the proposed system on PH2 database and 94.50% on international Skin Imaging Collaboration (ISIC) 2017 database.

关键词： Wavelet analysis Genetic programming Neural network Skin cancer classification Dermoscopic image categorization

来源：评论

学校读者我要写书评

暂无评论

Active Segmentation: Differential Geometry meets machine learning 22

Active Segmentation: Differential Geometry meets Machine Lea...

引用

23rd international conference on Computer Systems and Technologies, CompSysTech 2022

作者： Prodanov, Dimiter Vohra, Sumit Kumar Neuroscience Research Flanders Imec Leuven Belgium Bulgarian Academy of Sciences Sofia Bulgaria Berlin Germany

ISBN: (纸本)9781450396448

Image segmentation is an active area of research for more than 30 years. Traditional image segmentation algorithms are problem-specific and limited in scope. On the other hand, machine learning offers an alternative paradigm where predefined features are combined into different classifiers, providing pixel-level classification and segmentation. However, machine learning only can not address the question as to which features are appropriate for a certain classification problem. This paper presents a project supported in part by the international Neuroinformatics Coordination Facility through the Google Summer of code. The project resulted in an automated image segmentation and classification platform, called Active Segmentation for ImageJ (AS/IJ). The platform integrates a set of filters computing differential geometrical invariants and combines them with machine learning approaches. © 2022 ACM.

关键词： Support vector machines

来源：评论

学校读者我要写书评

暂无评论

Discovered changes in rice occupation with satellite images based on random forest approach 2020

Discovered changes in rice occupation with satellite images ...

引用

4th international conference on machine learning and soft computing (icmlsc)

作者： Hiep Xuan Huynh Ky Minh Nguyen Khoa Duc Nguyen Huong Hoang Luong Tran, Nghi C. Linh Thuy Thi Nguyen Tan Minh Tran Phuong Truc Thi Pham Simona Niculescu Can Tho Univ Coll Informat & Commun Technol Can Tho City Vietnam Fsoft Can Tho Stock Co Can Tho City Vietnam FPT Univ Can Tho City Vietnam Natl Cent Univ Taoyuan Taiwan Inst Univ Europeen Mer Plouzane France

ISBN: (纸本)9781450376310

Although agricultural production contributed a significant share of Vietnam's total production, the advancement and proficiency of remote sensing are still narrowly applied in this sector. Recent years, by the open access to satellite products of sufficient characteristics, the agriculture with satellite images supporting is being boosted. This paper focuses on identifying land use and its long-term changes in the selected regions of the Mekong Delta.

关键词： Rice satellite images remote sensing regions of interest random forest classification machine learning.

来源：评论

学校读者我要写书评

暂无评论

Intrinsic Imaging Model Enhanced Contrastive Face Representation learning 17

Intrinsic Imaging Model Enhanced Contrastive Face Representa...

引用

17th IEEE international conference on Automatic Face and Gesture Recognition, FG 2023

作者： Sun, Haomiao Shan, Shiguang Han, Hu Institute of Computing Technology Cas Beijing100190 China University of Chinese Academy of Sciences Beijing100049 China Pengcheng Laboratory Shenzhen518055 China

ISBN: (纸本)9798350345445

Humans can easily perceive numerous information from faces, only part of which has been achieved by a machine, thanks to the availability of large-scale face images with supervision signals of those specific tasks. More face perception tasks, like rare expression or attribute recognition, and genetic syndrome diagnosis, are not solved due to a critical shortage of supervised data. One possible way to solve these tasks is leveraging ubiquitous large-scale unsupervised face images and building a foundation face model via methods like contrastive learning (CL), which is, however, not aware of the intrinsic physics of the human face. In consideration of this shortcoming, this paper proposes to enhance contrastive face representation learning by the physical imaging model. Specifically, besides the CL-backbone network, we also design an auxiliary bypass pathway to constrain the CL-backbone to support the ability of accurately re-rendering the face with a differentiable physical imaging model after decomposing an input face image into intrinsic 3D imaging factors. With this design, the CL network is endowed the capacity of implicitly 'knowing' the 3D of the face rather than the 2D pixels only. In experiments, we learn face representations from the CelebA and WebFace-42M datasets in unsupervised mode and evaluate the generalization capability of the representations with three different downstream tasks in case of limited supervised data. The experimental results clearly justify the effectiveness of the proposed method. © 2023 IEEE.

关键词： learning systems

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：