检索结果-内蒙古大学图书馆

International Conference on science Technology Engineering and Management (ICONSTEM)

作者： Murali N D. Beulah David Department of Computer Science and Engineering Saveetha School of Engineering SIMATS Chennai Tamilnadu India Department of Data Analytics Institute of Information Technology Saveetha School of Engineering SIMATS Chennai Tamilnadu India

The safety of the vehicles is given great importance by advanced intelligent transportation systems. We are here to make sure that swift transit vehicles and infrastructure communicate with one other quickly. The majority of the vehicles need regular monitoring for appropriate warnings and road maps to achieve their ultimate objective in a timely manner in the hill station bad environment, and travel in hairpin bends and risky turning in particular is a duty that is vital for ITS. Here, my study will provide a strategy for preventing accidents by applying a sophisticated early warning system before notifying the car in a certain circumstance. In order to prevent accidents, machine learning is a technique where the system automatically learns and enhances the rapid object detection stage in bobby pin road turning scenarios. Here, my study offers effective unsupervised learning of these characteristics based on the acquired data, which in turn serves as the foundation for the clustering I performed, which successfully builds a mobile ad hoc network. In this work, we suggested an innovative automated discovering vehicle prier that uses the UNetXST technique in real-time to notify warring for vehicle in three-way shining LED light with blinking mode, slowing warring message in display unit, and beep alarm sound. By combining V2I and V2V technologies and sending messages to the central traffic light management system, we highlighted safety and communication. In worst-case scenarios, we employ a v2v technology sensor to detect the vehicle that will cross the junction road, and we investigate two-way merging technology's potential to implement and resolve these problems in transparent object tracking in the turning system. We looked at future direction and hurdles at this time in addition to a case study illustrating a VANET-based scenario of a critical urban road turning, receiving the earliest notice to flee from a traffic collision, and experiencing the perfect EV momen

关键词：

来源：评论

学校读者我要写书评

暂无评论

Hyperparameter Tuning and Model Evaluation in Causal Effect Estimation

arXiv

引用

arXiv 2023年

作者： Machlanski, Damian Samothrakis, Spyridon Clarke, Paul Department of Computer Science and Electronic Engineering University of Essex United Kingdom Institute for Analytics and Data Science University of Essex United Kingdom Institute for Social and Economic Research University of Essex United Kingdom

The performance of most causal effect estimators relies on accurate predictions of high-dimensional non-linear functions of the observed data. The remarkable flexibility of modern Machine Learning (ML) methods is perfectly suited to this task. However, data-driven hyperparameter tuning of ML methods requires effective model evaluation to avoid large errors in causal estimates, a task made more challenging because causal inference involves unavailable counterfactuals. Multiple performance-validation metrics have recently been proposed such that practitioners now not only have to make complex decisions about which causal estimators, ML learners and hyperparameters to choose, but also about which evaluation metric to use. This paper, motivated by unclear recommendations, investigates the interplay between the four different aspects of model evaluation for causal effect estimation. We develop a comprehensive experimental setup that involves many commonly used causal estimators, ML methods and evaluation approaches and apply it to four well-known causal inference benchmark datasets. Our results suggest that optimal hyperparameter tuning of ML learners is enough to reach state-of-the-art performance in effect estimation, regardless of estimators and learners. We conclude that most causal estimators are roughly equivalent in performance if tuned thoroughly enough. We also find hyperparameter tuning and model evaluation are much more important than causal estimators and ML methods. Finally, from the significant gap we find in estimation performance of popular evaluation metrics compared with optimal model selection choices, we call for more research into causal model evaluation to unlock the optimum performance not currently being delivered even by state-of-the-art procedures. © 2023, CC BY.

关键词： Petroleum reservoir evaluation

来源：评论

学校读者我要写书评

暂无评论

A Machine Learning Edged Self-Adaptable Vehicle Slowdown Earliest Warning Information System Using IOT Based Traffic Prediction Model

A Machine Learning Edged Self-Adaptable Vehicle Slowdown Ear...

引用

Artificial Intelligence and Knowledge Discovery in Concurrent Engineering (ICECONF), International Conference on

作者： Murali N D. Beulah David Department of Computer Science and Engineering SIMATS Saveetha School of Engineering Chennai Tamilnadu India Department of Data Analytics SIMATS Institute of Information Technology Saveetha School of Engineering Chennai Tamilnadu India

Development of vehicle safety is the work's primary goal, according to its abstract vehicle in traffic congestion using the Internet of Things (IoT). The intense growth of the city's vehicle population needs intelligent traffic systems to be considered resourcefully and sustainably by enhancing the full advantage of recent technology. The unpredictable traffic flow is a substantial issue that carries a colossal traffic movement in real-time scenarios. The random traffic control systems during peak and non-peak hours with unproductive human possessions, which leads to increased traffic and road offenses due to ineffective traffic monitoring systems. To sort down this dynamic scenario, the self-adaptable machine learning approach has been adopted to sequence the intelligent traffic flow by using IOT based early warning system in vehicles. The proposed work focused on the traffic congestion prediction operationalized using the unsupervised algorithm to train the gathered data sets using a neural network. It also aims to deliver a clarification that will upturn the comfort level of travelers to make intelligent and better transference choices. A neural network is a reasonable approach to finding traffic circumstances in sequence with this, the machine learning algorithms and their accuracy, which practices an outline in the collected data sets and then produces crucial decisions in evidence about traffic flow and congestion levels. The most focused part of the research is to enhance the intelligence in the IoT docking system, which prevents the vehicle congestion and returns a flawless, innovative design.

关键词： Machine learning algorithms Urban areas Sociology Neural networks Machine learning Traffic control Internet of Things

来源：评论

学校读者我要写书评

暂无评论

Efficiency Ordering of Stochastic Gradient Descent

arXiv

引用

arXiv 2022年

作者： Hu, Jie Doshi, Vishwaraj Eun, Do Young Department of Electrical and Computer Engineering North Carolina State University United States Data Science and Advanced Analytics IQVIA

We consider the stochastic gradient descent (SGD) algorithm driven by a general stochastic sequence, including i.i.d noise and random walk on an arbitrary graph, among others;and analyze it in the asymptotic sense. Specifically, we employ the notion of ‘efficiency ordering’, a well-analyzed tool for comparing the performance of Markov Chain Monte Carlo (MCMC) samplers, for SGD algorithms in the form of Loewner ordering of covariance matrices associated with the scaled iterate errors in the long term. Using this ordering, we show that input sequences that are more efficient for MCMC sampling also lead to smaller covariance of the errors for SGD algorithms in the limit. This also suggests that an arbitrarily weighted MSE of SGD iterates in the limit becomes smaller when driven by more efficient chains. Our finding is of particular interest in applications such as decentralized optimization and swarm learning, where SGD is implemented in a random walk fashion on the underlying communication graph for cost issues and/or data privacy. We demonstrate how certain non-Markovian processes, for which typical mixing-time based non-asymptotic bounds are intractable, can outperform their Markovian counterparts in the sense of efficiency ordering for SGD. We show the utility of our method by applying it to gradient descent with shuffling and mini-batch gradient descent, reaffirming key results from existing literature under a unified framework. Empirically, we also observe efficiency ordering for variants of SGD such as accelerated SGD and Adam, open up the possibility of extending our notion of efficiency ordering to a broader family of stochastic optimization algorithms. Copyright © 2022, The Authors. All rights reserved.

关键词： Efficiency

来源：评论

学校读者我要写书评

暂无评论

Robustness of Algorithms for Causal Structure Learning to Hyperparameter Choice

arXiv

引用

arXiv 2023年

Hyperparameters play a critical role in machine learning. Hyperparameter tuning can make the difference between state-of-the-art and poor prediction performance for any algorithm, but it is particularly challenging for structure learning due to its unsupervised nature. As a result, hyperparameter tuning is often neglected in favour of using the default values provided by a particular implementation of an algorithm. While there have been numerous studies on performance evaluation of causal discovery algorithms, how hyperparameters affect individual algorithms, as well as the choice of the best algorithm for a specific problem, has not been studied in depth before. This work addresses this gap by investigating the influence of hyperparameters on causal structure learning tasks. Specifically, we perform an empirical evaluation of hyperparameter selection for some seminal learning algorithms on datasets of varying levels of complexity. We find that, while the choice of algorithm remains crucial to obtaining state-of-the-art performance, hyperparameter selection in ensemble settings strongly influences the choice of algorithm, in that a poor choice of hyperparameters can lead to analysts using algorithms which do not give state-of-the-art performance for their data. © 2023, CC BY.

关键词： Petroleum reservoir evaluation

来源：评论

学校读者我要写书评

暂无评论

Cosmological parameter estimation and inference using deep summaries

引用

Physical Review D 2021年第12期104卷 123526-123526页

作者： Janis Fluri Tomasz Kacprzak Alexandre Refregier Aurelien Lucchi Thomas Hofmann Data Analytics Lab Department of Computer Science ETH Zurich 8006 Switzerland

The ability to obtain reliable point estimates of model parameters is of crucial importance in many fields of physics. This is often a difficult task given that the observed data can have a very high number of dimensions. In order to address this problem, we propose a novel approach to construct parameter estimators with a quantifiable bias using an order expansion of highly compressed deep summary statistics of the observed data. These summary statistics are learned automatically using an information maximising loss. Given an observation, we further show how one can use the constructed estimators to obtain approximate Bayes computation (ABC) posterior estimates and their corresponding uncertainties that can be used for parameter inference using Gaussian process regression even if the likelihood is not tractable. We validate our method with an application to the problem of cosmological parameter inference of weak lensing mass maps. We show in that case that the constructed estimators are unbiased and have an almost optimal variance, while the posterior distribution obtained with the Gaussian process regression is close to the true posterior and performs better or equally well than comparable methods.

关键词： Cosmological parameters Dark energy Dark matter

来源：评论

学校读者我要写书评

暂无评论

Predicting Electricity Consumption Using Load Profiling Clustering Method

Predicting Electricity Consumption Using Load Profiling Clus...

引用

data analytics for Business and Industry: Way Towards a Sustainable Economy (ICDABI), International Conference on

作者： Esra Al-Jowder Sawsan Hilal Riadh Ksantini M.Sc. in Big Data Science & Analytics Collage of Science University of Bahrain Kingdom of Bahrain Department of Mathematics College of Science University of Bahrain Kingdom of Bahrain Department of Computer Science College of Information Technology University of Bahrain Kindgom of Bahrain

ISBN: (数字)9798350369786

ISBN: (纸本)9798350369793

It became crucial to analyze the energy consumption trends from smart meter data to identify potential saving opportunities by the electricity suppliers and authorities since energy consumption is considered one of the essential topics of energy systems which gained high attention after the energy crisis in 1970’s. Despite that smart meter data analytics is relatively recent field, the progress in high performance computing makes it possible to efficiently and effectively extracting more beneficial knowledge and patterns from collected massive data within the smart grid. This research aims at providing a model for predicting load profiling while considering the temporal feature of the smart meter data by deploying a deep clustering algorithm embedded with an autoencoder and compare it with the dynamic autoencoder which can improve the prediction quality while applying dimensionality reduction and evaluate these models based on the performance using the silhouette score. The dynamic autoencoder outperformed the deep embedded clustering (DEC) and improved deep embedded clustering (IDEC) models and achieved a silhouette score of 0.9931 compared to DEC of 0.6128 and IDEC of 0.8829. This shows that positioning a dynamic autoencoder embedded with deep clustering model has worked effectively against the problem of feature drift and feature randomness.

关键词： Energy consumption data analysis Heuristic algorithms Electricity Clustering algorithms Predictive models Smart meters

来源：评论

学校读者我要写书评

暂无评论

Exploring Decentralized Identity Verification Systems Using Blockchain Technology: Opportunities and Challenges

Exploring Decentralized Identity Verification Systems Using ...

引用

Global Conference for Advancement in Technology (GCAT)

作者： V. Saravanan Sumalatha A Desidi Narsimha Reddy B. Shamreen Ahamed Udayakumar K Arthi Priyadharshini M Department of Nano Electronics Materials and Sensors Saveetha School of Engineering Saveetha Institute of Medical and Technical Sciences Saveetha University Chennai Tamilnadu India Department of Computer Science Kristu Jayanti College Autonomous Bengaluru Data Consultant (Data Governance Data Analytics: Enterprise Performance Management AI&ML) Soniks Consulting LLC Texas USA Department of Computer Science and Engineering Sathyabama Institute of Science and Technology Jeppiaar Nagar Chennai Department of Computer Science and Engineering Vel Tech Rangarajan Dr. Sagunthala R&D Institute of Science and Technology India Department of CSE Dhanalakshmi College of Engineering Chennai

ISBN: (数字)9798350376685

ISBN: (纸本)9798350376692

The rapid advancement of digital technologies has underscored the critical necessity for robust and reliable identity verification systems in today's interconnected world. Traditional centralized identity management systems encounter myriad challenges, from vulnerabilities to data breaches and privacy concerns, to inefficiencies inherent in verification processes. This paper delves into the transformative potential of decentralized identity verification systems using blockchain technology as a robust solution to these pressing issues. By harnessing blockchain's immutable, transparent, and secure characteristics, decentralized identity systems not only promise enhanced privacy, security, and user control over personal data but also foster a paradigm shift towards more efficient and user-centric identity management *** study rigorously examines the foundational principles of blockchain technology and its strategic application in decentralized identity verification. It highlights key opportunities such as strengthened security measures through cryptographic protocols, enhanced user autonomy in managing and sharing personal information, and streamlined verification processes that reduce operational complexities and costs. Additionally, the paper critically evaluates significant challenges that must be addressed, including scalability concerns as blockchain networks grow, complexities in navigating regulatory landscapes designed for centralized systems, and overcoming barriers to widespread user adoption. Through an exhaustive analysis of existing models, real-world case studies, and emerging trends, this paper aims to provide comprehensive insights into the feasibility and future prospects of decentralized identity verification systems. By contributing to the ongoing discourse on innovative identity management solutions in the digital age, this research seeks to pave the way for more secure, efficient, and equitable digital ecosystems.

关键词： data privacy Navigation Scalability Pressing Market research Blockchains Complexity theory Security Reliability Identity management systems

来源：评论

学校读者我要写书评

暂无评论

Development and Innovations in Wearable Sensors and Devices for Human Machine Interaction in Virtual Platform 2

Development and Innovations in Wearable Sensors and Devices ...

引用

2nd International Conference on Augmented Intelligence and Sustainable Systems, ICAISS 2023

作者： Singh, Rahul Javali, Anand Priya, S. Vardhana Naidu, B Vishnu Gautam, Gaurav Presidium Gurgaon Sec-57 Pgt Physics Department of Physics Haryana 122103 India Conduent Technology Services - Usa Data Analytics and Innovations Dept FriscoTX75035 United States Nehru Institute of Engineering and Technology Department of Computer Science and Engineering Tamilnadu Coimbatore India Mohan Babu University Department of Mechanical Engineering Andhra Pradesh Tirupati India Graphic Era Hill University Department of Physics Dehradun India

ISBN: (纸本)9798350325799

Recent advances in deep learning algorithms and wearable sensor technology have opened the door for creative methods to improve human-machine interaction (HMI) on virtual platforms. The use of wearable sensors and devices in conjunction with hybrid deep learning algorithms is presented in detail in this abstract. Real-time physiological and contextual data from users can be collected with the help of wearable sensors integrated into devices like smartwatches, augmented reality goggles, and biometric monitoring equipment. This information can then be used to improve augmented reality (AR) and virtual reality (VR) systems, enabling more effortless and intuitive user interactions. A seamless connection between the user and the virtual platform is made possible by the combination of sensor data and deep learning models, which enables the creation of complex algorithms that can decipher user intent, gestures, emotions, and physical actions. These hybrid models may efficiently learn complicated patterns from wearable sensor data by combining standard feature extraction techniques with convolutional neural networks (CNNs), recurrent neural networks (RNNs), and generative adversarial networks (GANs). Wearable sensors and hybrid deep learning techniques have numerous and significant applications in virtual platforms. They make it possible for, among other things, real-time activity recognition for individualized virtual fitness coaching, emotion recognition for adaptive and sympathetic virtual characters, and gesture recognition for intuitive control of virtual objects. In summary, combining wearable sensors and devices with hybrid deep learning algorithms has enormous potential to change how humans and machines interact on virtual platforms. © 2023 IEEE.

关键词： Emotion Recognition

来源：评论

学校读者我要写书评

暂无评论

Cutting-Edge Robotic Arm Design: Harnessing Flex Sensors and Novel Controller Framework to Meet Real-World Innovations

Cutting-Edge Robotic Arm Design: Harnessing Flex Sensors and...

引用

2024 International Conference on Intelligent Systems for Cybersecurity, ISCS 2024

作者： Syambabu, V. Kalaiselvi, K. Mahil, S. Derisha Hemalatha, S. Alekhya, Badi Kanagavalli, N. Mlr Institute of Technology Department of Electronics and Communication Engineering Hyderabad India Saveetha College of Liberal Arts and Sciences Saveetha Institute of Medical and Technical Sciences Department of Data Analytics Tamil Nadu Chennai India Tamil Nadu Chennai India Prince Shri Venkateshwara Padmavathy Engineering College Department of Electrical and Electronics Engineering Tamil Nadu Chennai India R.M.D Engineering College Department of Aiml Tamil Nadu Chennai India Rajalakshmi Institute of Technology Department of Computer Science and Engineering Tamil Nadu Chennai India

ISBN: (纸本)9798350375237

This research paper introduces a sophisticated arm movement tracking system, leveraging a multi-sensor fusion approach for enhanced precision and versatility. The comprehensive system comprises various components, including a dedicated cabinet, six Leap Motion sensors, six microcomputers, a Kinect depth camera, a KVM switcher, a router, and flex sensors. In addition to the aforementioned components, the integration of flex sensors enhances the system's capability to capture nuanced movements and gestures of the human arm. Flex sensors contribute valuable data to the multi-sensor fusion, further refining the arm movement tracking and providing a comprehensive understanding of the intricacies involved. By quantifying the Root Mean Square Error (RMSE) for each filtering technique, including the data from flex sensors, we gained valuable insights into the accuracy and precision achieved by Particle Filtering in refining data from the specified angles. This approach ensures a thorough validation of Particle Filtering's superior performance when compared to Existing filtering Methods. The systematic evaluation of RMSE serves as a robust method for assessing the effectiveness of Particle Filtering, affirming its potential as a high-performing solution for precise angle estimation and tracking within the context of the presented arm movement tracking system that incorporates flex sensors. © 2024 IEEE.

关键词： Cameras

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：