检索结果-内蒙古大学图书馆

An Actor-Critic Method for Simulation-Based Optimization

IFAC-PapersOnLine 2022年第11期55卷 7-12页

作者： Kuo Li Qing-Shan Jia Jiaqi Yan Dept. of Automation BNRist Tsinghua University Beijing China Dept. of Computer Science Tokyo Institute of Technology Japan

In this work, we study simulation-based optimization, where the agent aims to select the best configuration from the design space with as few as possible iterations. Inspired by the success of deep reinforcement learning (DRL), we formulate the sampling process as policy searching and give a solving method from the perspective of policy iteration. Concretely, a surrogate model for predicting the performance of each configuration and a parameterized sampling policy are applied, which correspond to the critic and actor in actor-critic (AC) method, respectively. We further derive the updating rule and propose two algorithms for configuration selection in continuous and discrete design spaces, respectively. Finally, the algorithms are validated experimentally on 1) two toy examples to intuitively explain the principle and 2) two high-dimensional tasks to reveal the effectiveness in large-scale problems. The results show that the proposed algorithms can efficiently deal with large-scale problems and effectively eliminate sub-optimal configurations.

关键词： simulation-based optimization reinforcement learning policy searching actor-critic policy iteration

来源：评论

学校读者我要写书评

暂无评论

Analysis of Defect Associated with Powder Bed Fusion with Deep Learning and Explainable AI 15

Analysis of Defect Associated with Powder Bed Fusion with De...

引用

15th International Conference on Knowledge and Smart technology, KST 2023

作者： Pratap, Ayush Sardana, Neha Utomo, Sapdo John, A. Karthikeyan, P. Hsiung, Pao-Ann Indian Institute of Technology Dept. of Metallurgical and Materials Engineering Ropar Rupnagar India National Chung Cheng University Grad. Ins. Ambient Intelligence and Smart Systems Chiayi County Taiwan National Chung Cheng University Dept. of Computer Science and Information Engineering Chiayi County Taiwan

ISBN: (纸本)9781665477123

Research into the detection, classification, and prediction of internal defects using surface morphology data of parts created via powder bed fusion-type additive manufacturing has become a hot topic in the previous decade thanks to the development of deep learning. However, there is no other evidence to evaluate the model other than accuracy and metrics. In this paper, a novel data set is compiled from various literature and other sources to evaluate the black box model using explainable artificial intelligence (XAI). The data set contains three major powder bed fusion defects: gas porosity, lack of fusion, and balling. The anomaly was initially found using convolutional neural networks (CNN) and transfer learning. Based on test data, a model comparison was performed to determine the best accuracy and an F1 score. VGG16 has outperformed all other models in terms of accuracy, with an F1 score of 98.6 percent. Further, the model has been compared with the existing state-of-the-art model for classification in the domain of powder bed fusion defects. Finally, VGG16 was employed to interpret and explain the test data set. The LIME explanations revealed that the feature predicted by the model was present in conjunction with the fault. As a result, we are confident that the proposed model with XAI would considerably improve the fairness and trustworthiness of the output result in the powder bed fusion field. This can also aid in the automation of additive manufacturing in the realm of Industry 4.0. © 2023 IEEE.

关键词： computer vision

来源：评论

学校读者我要写书评

暂无评论

An Open Hyperspectral Dataset with Sea-Land-Cloud Ground-Truth from the Hypso-1 Satellite

An Open Hyperspectral Dataset with Sea-Land-Cloud Ground-Tru...

引用

Workshop on Hyperspectral Image and Signal Processing: Evolution in Remote Sensing, WHISPERS

作者： Jon A. Justo Joseph Garrett Dennis D. Langer Marie B. Henriksen Radu T. Ionescu Tor A. Johansen Dept. Engineering Cybernetics Norwegian University of Science and Technology Trondheim Norway Dept. Marine Technology Norwegian University of Science and Technology Trondheim Norway Dept. of Computer Science University of Bucharest Romania

Hyperspectral Imaging, employed in satellites for space remote sensing, like HYPSO-1, faces constraints due to few labeled data sets, affecting the training of AI models demanding these ground-truth annotations. In this work, we introduce The HYPSO-1 Sea-Land-Cloud-Labeled Dataset, an open dataset with 200 diverse hyperspectral images from the HYPSO-1 mission, available in both raw and calibrated forms for scientific research in Earth observation. Moreover, 38 of these images from different countries include ground-truth labels at pixel-level totaling about 25 million spectral signatures labeled for sea/land/cloud categories. To demonstrate the potential of the dataset and its labeled subset, we have additionally optimized a deep learning model (1D Fully Convolutional Network), achieving superior performance to the current state of the art. The complete dataset, ground-truth labels, deep learning model, and software code are openly accessible for download at the website https://***/hypso1_sea_land_clouds_dataset/.

关键词：

来源：评论

学校读者我要写书评

暂无评论

AN OPEN HYPERSPECTRAL DATASET WITH SEA-LAND-CLOUD GROUND-TRUTH FROM THE HYPSO-1 SATELLITE

arXiv

引用

arXiv 2023年

作者： Justo, Jon A. Garrett, Joseph Langer, Dennis D. Henriksen, Marie B. Ionescu, Radu T. Johansen, Tor A. Dept. Engineering Cybernetics Norwegian University of Science and Technology Trondheim Norway Dept. Marine Technology Norwegian University of Science and Technology Trondheim Norway Dept. of Computer Science University of Bucharest Romania

关键词： Remote sensing

来源：评论

学校读者我要写书评

暂无评论

Business intelligence based novel marketing strategy approach using automatic speech recognition and text summarization 2

Business intelligence based novel marketing strategy approac...

引用

2nd International Conference on Computing and Data Science, CDS 2021

作者： Hegdepatil, Pratidnya Davuluri, Kiran NMIMS-MPSTME Dept. of Information Technology Mumbai India IIITDM Kancheepuram Dept. of Computer Science Chennai India

ISBN: (纸本)9781665404280

A novel marketing strategy approach is proposed that offers distinctive features than other competitive models. Text converted from speech between customer and sales executive is fed into text summarizer to automate customer strategy into various priorities. The approach shows that there is no need of data entry for emulating marketing tools. The need of account for customer's preferences is high, which can be stored by capturing the customer's context. Data mining and advanced analytical aspects give accurate results to help marketing managers and improve firms with solid customer strategies. The experimental results for the implementation show high outcomes in comparison to native advances in the technologies. A total of 62 test cases have been performed where an average of 12% Word Error Rate has been observed for Speech Recognition, which is optimal than other standard Speech Datasets. Furthermore, summarization compressed rate has been reported as 60%. The automated categorization module has achieved 70 % accuracy and overall F-score is 0.81349. © 2021 IEEE.

关键词： Speech recognition

来源：评论

学校读者我要写书评

暂无评论

Enhancing Hyperspectral Image Compression Through Stacked Autoencoder Approach

Enhancing Hyperspectral Image Compression Through Stacked Au...

引用

International Conference on Electrical Engineering and Information & Communication technology (ICEEICT)

作者： Afsana Afrin Md. Rakibul Haque Md. Al Mamun Institute of Information and Communication Technology Rajshahi University of Engineering & Technology Rajshahi Bangladesh Dept. of Computer Science & Engineering Rajshahi University of Engineering & Technology Rajshahi Bangladesh

ISBN: (数字)9798350385779

ISBN: (纸本)9798350385786

Hyperspectral Image (HSI) consisting of numerous high-resolution spectral bands creates challenges of the high dimensionality problem in HSI classification which hinder real-life applications despite their abundance of information. We present a lossy compression approach using stacked autoencoders to reduce high dimensionality problem in this paper. The pro-posed method utilizes stacked autoencoders to extract features from HSIs, allowing compression and subsequent reconstruction. The study demonstrates improved Peak Signal-to-Noise Ratio (PSNR) of 70.43, 60.87 and 61.36 on three distinct HSI datasets-Salinas, Botswana, and KSC respectively compared to previous autoencoder-based compression method. Additionally, the impact of compression on classification accuracy is assessed using a 3D-2D CNN model, achieving an average accuracy of 99 % across the datasets. Improved compression along with high classification accuracy shows our proposed method's usefulness.

关键词： Image coding PSNR Feature extraction Communications technology Image reconstruction Hyperspectral imaging

来源：评论

学校读者我要写书评

暂无评论

Design of an Efficient Model for Enhanced IoT Security and Maintenance in Smart Cities

Design of an Efficient Model for Enhanced IoT Security and M...

引用

Technological Innovations and Advance Computing (TIACOMP), International Conference on

作者： Saurav Bhandari Sumit Chaudhary Himadri Vaidya Ujjwal Kumar Madhu Kirola Harishchander Anandaram Dept. of Computer Science and Engineering Uttaranchal Institute of Technology Uttaranchal University Dehradun India Dept. of Computer Science and Engineering Graphic Era Hill University Graphic Era Deemed to be University Dehradun India Dept. of Mechanical Engineering JB Institute of Technology Dehradun India Amrita School of Artificial Intelligence Coimbatore Amrita Vishwa Vidyapeetham India

ISBN: (数字)9798350392111

ISBN: (纸本)9798350392128

It is of the utmost importance to ensure robust security and effective device maintenance in the rapidly changing landscape of smart city infrastructures. Be that as it may, existing procedures frequently miss the mark in tending to the dynamic and different difficulties presented by emerging dangers and the perplexing exchange of IoT environments. This paper presents a clever way to deal with these issues through three inventive strategies: Adaptive Reinforcement Learning for IoT Security (ARLIS), Graph-Analytics-Based Anomaly Detection (GABAD), and Predictive Maintenance using Time Series Forecasting (PMTSF). Customary safety efforts frequently battle to adjust to the advancing danger scene, prompting weaknesses and likely breaks. ARLIS tends to this constraint by loading the force of support figuring out how to progressively change security conventions in view of ongoing IoT information and organization conditions. ARLIS continuously learns from feedback to improve security measures, which reduces response times to security threats and improves overall security efficacy. Also, irregularity location in IoT networks is tested by the sheer volume and complexity of information produced by interconnected gadgets. GABAD beats this obstacle by utilizing diagram examination to recognize oddities in device connections and ways of behaving inside smart city organizations. By building an exhaustive chart model and applying progressed oddity identification calculations, GABAD accomplishes unequaled precision in peculiarity location while limiting misleading up-sides. In addition, optimizing system reliability and minimizing service disruptions require proactive maintenance of IoT devices. PMTSF alters upkeep methodologies by utilizing time series to foresee gadget disappointments before they happen. By examining gadget execution measurements and ecological variables, PMTSF empowers convenient support mediations, prompting a significant decrease in spontaneous margin time and

关键词： Technological innovation Smart cities Biological system modeling Time series analysis Organizations Time measurement Maintenance Security Internet of Things Time factors

来源：评论

学校读者我要写书评

暂无评论

Evaluation of Digital Wallet Transaction Accuracy using Machine Learning 2

Evaluation of Digital Wallet Transaction Accuracy using Mach...

引用

2nd International Conference on Advance Computing and Innovative Technologies in Engineering, ICACITE 2022

作者： Pallivalappil, Abdul Shareef Sinha, Divyanshu Kumari, P. Lalitha Katti, Anvesha Sakthi, G. Narkhede, Mayur Ravindra College of Computer Science and Information Science Srinivas University Mangalore India Uttarakhand Technical University Dept. of CSE Dehradun India Visakha Institute of Engineering & Technology Dept. of CSE AP Visakhapatnam India The NorthCap University Dept. of CSE Haryana Gurgaon India Hindusthan College of Engineering and Technology Dept. of CSE TN Coimbatore India Vishwakarma Institute of Technology Dept. of Ai Coe Pune India

ISBN: (纸本)9781665437899

Non-card retail payments, such as direct-from-the-bank programs and E-wallets, have become increasingly common at both brick-and-mortar and online establishments. In recent years, the use of mobile wallets has increased. It relates to any form of computer program that can 'learn' on its own beyond the intervention of a human. Machine learning can detect out-of-the-ordinary purchases and flag them as fraudulent by continuously monitoring data in real-time. Machine learning models are being improved by developing new methodologies based on pre-existing models. An example is a hybrid model in which the dataset is under-sampled and over-sampled;for example, credit card transactions. Digital wallets and machine learning are the main focus of this article. © 2022 IEEE.

关键词： Deep learning

来源：评论

学校读者我要写书评

暂无评论

Data Security Enhancement in Cloud Computing by Proposing A DKE Encryption Protocol 23

Data Security Enhancement in Cloud Computing by Proposing A ...

引用

Proceedings of the 2023 International Conference on Advances in Artificial Intelligence and Applications

作者： Sahar Ebadinezhad Halmat Ayub Abdulmajed Department of Computer Information Systems Near East University North Cyprus and Computer Information Systems Research and Technology Center (CISRT) Near East University North Cyprus Dept. Engineering and Computer Systems Lebanese French University (LFU) Iraq

ISBN: (纸本)9798400708268

This study uses cryptography to tackle the important problem of data security in cloud computing. Two keys are used in the Dual Key Encryption (DKE) method for both encryption and decryption. DKE employs a public key for the first encryption round before cloud upload and a user-only private key for the second round of encryption. Decryption operates oppositely. When tested and simulated with different file sizes on a CloudAnalyst simulator, DKE operates faster and more efficiently than cryptographic methods such as Triple DES (3DES), AES, RSA, and DES. Thanks to developments in information technology, cloud computing has made several services available online. Data security assurance is still very difficult to achieve, nevertheless. In this context, cryptography is essential and has led to the creation of the novel DKE technology. This technology is not only far more efficient than well-known cryptographic techniques such as 3DES, but it also improves data security.

关键词： Cloud Services Cryptography Data Integrity Data Security Dual Key Encryption

来源：评论

学校读者我要写书评

暂无评论

Simplified Micropayment Mechanism to Eliminate the Risk of Double Payment in E-Commerce

Simplified Micropayment Mechanism to Eliminate the Risk of D...

引用

Advances in Intelligent Computing and Applications (AICAPS), International Conference on

作者： Narendra Kumar Chahar Krishan Pal Singh Muzzammil Hussain Dept. of Data Science and Analytics Central University of Rajasthan Ajmer India Dept. of Computer Science and Engineering Malaviya National Institute of Technology Jaipur India Dept. of Computer Science and Engineering Central University of Rajasthan Ajmer India

In recent years, the emergence of the Internet and E-commerce has steered significant growth in digital transactions. Businesses today need mobile wallets, credit and debit cards, and e-cash to digitize payments. Digital payment systems are in transition and promise amazing advancements, but they also pose many risks and as the number of online transactions is increasing tremendously, we need a security system that follows all security norms. In this paper, we study digital transaction systems and evaluate various components of E-commerce plat-forms to address the security of these services. We evaluate the attributes that affect the security of digital payment processes and identify several barriers that hinder their performance to propose a simplified payment mechanism for micro-payments that eliminates the double payment problem.

关键词： Micropayments Internet Security Electronic commerce Business

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：