检索结果-内蒙古大学图书馆

11th International Conference on Learning Representations, ICLR 2023

作者： Shu, Yao Dai, Zhongxiang Sng, Weicong Verma, Arun Jaillet, Patrick Low, Bryan Kian Hsiang Dept.of Computer Science National University of Singapore Singapore Dept.of Electrical Engineering and Computer Science MIT United States

Zeroth-order (ZO) optimization, in which the derivative is unavailable, has recently succeeded in many important machine learning *** algorithms rely on finite difference (FD) methods for derivative estimation and gradient descent (GD)-based approaches for ***, these algorithms suffer from query inefficiency because many additional function queries are required for derivative estimation in their every GD update, which typically hinders their deployment in real-world applications where every function query is *** this end, we propose a trajectory-informed derivative estimation method which only employs the optimization trajectory (i.e., the history of function queries during optimization) and hence can eliminate the need for additional function queries to estimate a ***, based on our derivative estimation, we propose the technique of dynamic virtual updates, which allows us to reliably perform multiple steps of GD updates without reapplying derivative *** on these two contributions, we introduce the zeroth-order optimization with trajectory-informed derivative estimation (ZORD) algorithm for query-efficient ZO *** theoretically demonstrate that our trajectory-informed derivative estimation and our ZORD algorithm improve over existing approaches, which is then supported by our real-world experiments such as black-box adversarial attack, non-differentiable metric optimization, and derivative-free reinforcement learning. © 2023 11th International Conference on Learning Representations, ICLR 2023. All rights reserved.

关键词： Trajectories

来源：评论

学校读者我要写书评

暂无评论

RFID-based 3D human pose tracking: A subject generalization approach

引用

Digital Communications and Networks 2022年第3期8卷 278-288页

作者： Chao Yang Xuyu Wang Shiwen Mao Dept.of Electrical and Computer Engineering Auburn UniversityAuburnAL36849-5201USA Dept.of Computer Science California State UniversitySacramentoCA95819-6021USA

Three-dimensional (3D) human pose tracking has recently attracted more and more attention in the computer vision field. Real-time pose tracking is highly useful in various domains such as video surveillance, somatosensory games, and human-computer interaction. However, vision-based pose tracking techniques usually raise privacy concerns, making human pose tracking without vision data usage an important problem. Thus, we propose using Radio Frequency Identification (RFID) as a pose tracking technique via a low-cost wearable sensing device. Although our prior work illustrated how deep learning could transfer RFID data into real-time human poses, generalization for different subjects remains challenging. This paper proposes a subject-adaptive technique to address this generalization problem. In the proposed system, termed Cycle-Pose, we leverage a cross-skeleton learning structure to improve the adaptability of the deep learning model to different human skeletons. Moreover, our novel cycle kinematic network is proposed for unpaired RFID and labeled pose data from different subjects. The Cycle-Pose system is implemented and evaluated by comparing its prototype with a traditional RFID pose tracking system. The experimental results demonstrate that Cycle-Pose can achieve lower estimation error and better subject generalization than the traditional system.

关键词： Radio-frequency identification(RFID) Three-dimensional(3D)human pose tracking Cycle-consistent adversarial network Generalization

来源：评论

学校读者我要写书评

暂无评论

E-ATM For Card Less Transactions Using Machine Learning 1

E-ATM For Card Less Transactions Using Machine Learning

引用

1st IEEE International Conference on Electronics, Computing, Communication and Control Technology, ICECCC 2024

作者： Aishwarya, D. Muralikumar, A. Mohammed Inthiyaz, M. Sathyabama Institute of Science and Technology Dept.of Computer Science and Engineering Tamilnadu Chennai600119 India

ISBN: (数字)9798350371802

ISBN: (纸本)9798350371802

A fingerprint recognition technique in the ATM system is the foundation of this study. We chose this to improve client security and facilitate simple transactions. Since every person has unique fingerprint details, this method will be more widely applicable. It is not necessary to keep your ATM card on hand at all times, nor is there any concern that your card will be lost. It is shown that fingerprint recognition technology outperforms and is safer than other technologies when compared to other ATM security systems already in use. This facilitates safe and simple transactions while maintaining a user-friendly environment. Without a doubt, service providers should also consider security when designing their ATM products, in order to protect financial transactions against security threats. And a lot about security holes, identity theft, credit card fraud, and ways for hackers and criminals to get any sensible passwords or pin numbers. Similar to a username, identification is employed in the present architecture to provide access to the system. To ensure that the intended user is indeed who say they are, since passwords may be stolen or lost. Biometric identity and authentication technologies are the most modern answers to security and privacy problems. The study of traits that may be used to identify a person is known as face recognition. These bodily attributes include, among other things, faces, voices, fingerprints, and hand shapes. The different security system authentication techniques are surveyed in this paper. This face biometric with QR code verification offers the highest degree of security in authentication. One of the most promising technologies for electronic money transactions is this procedure. The need for quick and precise user identification and authentication has increased due to the expansion of the electronic transaction industry. The Automated Teller Machine (ATM) is a fantastic convenience for people to satisfy their financial requirements. Not

关键词： Authentication

来源：评论

学校读者我要写书评

暂无评论

Analysis of Pattern Matching Techniques used for Information Leakage 15

Analysis of Pattern Matching Techniques used for Information...

引用

15th International Conference on Advances in Computing, Control, and Telecommunication Technologies, ACT 2024

作者： Kumar, Abhishek Senapati, K.K. Patnaik, K.S. Dept.of Computer Science & Engineering Birla Institute of Technology Mesra Ranchi India

ISBN: (纸本)9798331300579

In the rapidly evolving digital landscape, ensuring the utmost security of sensitive information against the pervasive risks of information leakage has become a paramount *** study dives deep into a critical facet of information security-the intricate world of pattern matching algorithms-vital tools in the ongoing battle against Information *** research embarks on a comprehensive exploration of a diverse array of pattern matching algorithms, shedding light on their pivotal roles in the identification and effective mitigation of potential information *** meticulously analyze their inherent strengths, nuanced limitations, and practical applications across real-world contexts, understanding their multifaceted ***, we delve into the complexities of commonly utilized datasets for training and rigorous testing, strategically accentuating essential evaluation criteria that guide algorithmic *** this extensive exploration, we seek to elucidate the landscape of pattern matching algorithms, placing emphasis on their dynamic nature and promising potential in advancing information ***, we put forth insightful proposals for pective research directions within this vital domain, aiming to shape the future of information protection and fortification. © Grenze Scientific Society, 2024.

关键词： Information leakage

来源：评论

学校读者我要写书评

暂无评论

Enhancing Metaverse using Quantum Technologies

Enhancing Metaverse using Quantum Technologies

引用

2024 International Conference on Advances in Computing Research on science engineering and Technology, ACROSET 2024

作者： Doshi, Jash Shaikh, Sanober Sultana Thadomal Shahani Engineering College Dept.of Computer Engineering Mumbai India Thadomal Shahani Engineering College Dept.of Artifical Intelligence and Data Science Mumbai India

ISBN: (纸本)9798350388800

Recently, metaverse technology has seen a boom in users. However, data privacy, security, connectivity, and computational complexity are still the main issues that ruin the user experience and make the metaverse less likely to be used. In this context, we propose a 3-layer quantum-enabled architecture system for the metaverse that holds the promise of overcoming these issues. We also cover the applications of the metaverse, the current challenges faced by the metaverse, and the exciting potential of Quantum Computing to revolutionize the metaverse. © 2024 IEEE.

关键词： Quantum computers

来源：评论

学校读者我要写书评

暂无评论

Enhancing Cloud Storage Security with Multi-Authority Authentication and File Encryption 7

Enhancing Cloud Storage Security with Multi-Authority Authen...

引用

7th International Conference on Inventive Computation Technologies, ICICT 2024

作者： Jorsen Fredlin, J. Rahul Vijay, P. Janani, S. St.Joseph's College of Engineering Dept.of Computer Science And Engineering Chennai India

ISBN: (数字)9798350359299

ISBN: (纸本)9798350359299

The integration of a multi-authority system poses significant challenges but offers numerous benefits for effective collaboration and decision-making across departments or organizations. Harmonizing laws, regulations, and policies, as well as integrating diverse technologies and data sources, are key hurdles in this process, requiring thorough planning, coordination, and implementation. Active participation from relevant authorities is essential, facilitated by open communication, collaborative environments, and ongoing training. Security and privacy considerations must also be addressed to safeguard sensitive information and ensure compliance with legislation. Successfully integrated multi-authority systems yield advantages such as increased communication, operational efficiency, and decision-making skills, while encouraging innovation, streamlining procedures, and reducing redundancy. This study emphasizes the importance of cooperation, decision-making, coordination, data management, technology integration, and security in achieving the potential benefits of a unified multi-authority system. © 2024 IEEE.

关键词： Decision making

来源：评论

学校读者我要写书评

暂无评论

Moth flame optimization algorithm for efficient feature selection and hybrid classification technique for intrusion detection system

引用

Multimedia Tools and Applications 2025年 1-21页

作者： Keerthanarani, U. Selvakumar, V. Singaravelan, S. Raja, S. Edwin Thenmozhi, K. Shunmugam, D. Arun Dept. of Computer Science and Engineering P.S.R Engineering College Sivakasi India Dept.Of Computer Science and Engineering Vel Tech Rangarajan Dr Sagunthala R & d Institute of Science and Technology Chennai Avadi India Dept. Of Computer Science and Engineering Coimbatore Institute of Engineering and Technology Coimbatore India Dept. Of Electronics and Communication Engineering P.S.R Engineering College Sivakasi India

The rapid expansion of the Internet of Things (IoT) brings numerous benefits but further presents fresh difficulties, especially in terms of security. The distributed and interconnected nature of IoT devices makes them susceptible to various security threats. An Intrusion Detection System (IDS) plays a crucial role in safeguarding IoT networks by monitoring, detecting, and responding to potential security incidents. The Internet of Things (IoT) connects billions of devices, sensors, and systems, transforming industries and everyday life. However, this extensive interconnectivity also introduces significant security risks, making Intrusion Detection Systems (IDS) a critical component for safeguarding IoT environments. Feature selection is a crucial aspect of building effective Intrusion Detection Systems (IDS). Selecting relevant features not only helps in achieving better classification accuracy but also contributes to reducing false alarm rates. In the proposed system a moth flame optimizer is utilized to select the efficient features from the dataset, overall accuracy of over 97%. Further it is classified employing Random Forest and J48 classifier which attains high classification accuracy and low false alarm rate. © The Author(s), under exclusive licence to Springer science+Business Media, LLC, part of Springer Nature 2025.

关键词： Feature Selection

来源：评论

学校读者我要写书评

暂无评论

A Real-Time Worker's Monitoring System using Deep Learning 15

A Real-Time Worker's Monitoring System using Deep Learning

引用

15th International Conference on Advances in Computing, Control, and Telecommunication Technologies, ACT 2024

作者： Harshavardhan, L. Kumar, Girish T.P. Gowda, Jayanth B.R. Keerthan, B.K. Bevinatti, Mohan K.S. Dept.of Computer Science and Engineering BGS Institute of Technology Karnataka BG Nagara India

ISBN: (纸本)9798331300579

The Real-Time Worker Monitoring System employing deep learning and web integration represents a cutting-edge solution for enhancing workplace safety and *** advanced deep learning algorithms, this system continuously analyzes real-time data from various sensors, such as cameras and wearable devices, to monitor workers' activities and *** integration with web platforms enables instant access to the monitored information, facilitating seamless communication and *** providing timely insights into worker behavior and environmental factors, this innovative system not only promotes a safer work environment but also optimizes productivity through informed management *** the convergence of deep learning technology and web integration, this monitoring system stands at the forefront of workplace safety solutions, offering a robust framework for real-time insights and proactive intervention. © Grenze Scientific Society, 2024.

关键词： Decision making

来源：评论

学校读者我要写书评

暂无评论

Automated Unified System for LPG Refill Booking & Lekage Detection: A Pervasive Approach 15

Automated Unified System for LPG Refill Booking & Lekage Det...

引用

15th International Conference on Advances in Computing, Control, and Telecommunication Technologies, ACT 2024

作者： Suraksha, P. Aishwarya Kumari, Kavana L. Dept.of Computer Science and Engineering Vidyavardhaka College of Engineering Karnataka Mysuru India Dept.of Artificial Intelligence and Machine learning Vidyavardhaka College of Engineering Karnataka Mysuru India

ISBN: (纸本)9798331300579

Civilization is becoming increasingly aware of natural resources, it results in the use of propane and butane, and hence the use of *** India most people use the LPG *** per current government regulations, intense demand but shortage in production of LPG cylinder, once a new cylinder is booked we need to wait for some days to get it *** no cost-efficient technique is available which identifies when cylinder is about to get *** this paper we designed, developed, a proven Automated Unified System using smart Pervasive sensors, where LPG retailer requests for a new cylinder using Short Message Service (SMS) once it identifies cylinder will be empty in few days and also provides detection of gas leakage for better service to the consumers and reduce the scope of irregularities. © Grenze Scientific Society, 2024.

关键词： Liquefied petroleum gas

来源：评论

学校读者我要写书评

暂无评论

Automated Fire Detection System Using Image Surveillance System (ISS) and Convolutional Neural Networks (CNN) 9

Automated Fire Detection System Using Image Surveillance Sys...

引用

9th International Conference on Advanced Computing and Communication Systems, ICACCS 2023

作者： Nagababu, P. Dhakshitha, K. Chandrika, G. Chowdary, U. Rohith Dept.of Computer Science and Engineering Andhra Pradesh Mylavaram India

ISBN: (纸本)9798350397376

Fire has become more complicated for places such as forests, malls, industries, functions with huge crowds, and theaters. Fires cause massive losses to all areas every year. There are a lot of severe threats to the people, assets, environment, and human and animal life with these existing fire safety methods. Detecting the fire in the early stages helps prevent the loss of people, properties, plants, etc. In some places they will use the smoke detectors to find the seriousness of the fire. Deep learning algorithms are most widely used to find the abnormalities, such as firing at specific places. Many existing models can be used to find accurate fire detection using various online sources such as images and videos. Various limitations of existing models are identified that will reduce the accuracy. This paper describes the new fire detection model that detects fire at any place in the early stages. The proposed automated model combined with a high-end image surveillance system (ISS) and convolutional neural networks (CNN) achieved better accuracy of fire detection in images. Bounding box is one of the models that find the fire affected regions from the dataset input images. Evaluation results show that the performance of proposed system achieved the better results. © 2023 IEEE.

关键词： Fires

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：