检索结果-内蒙古大学图书馆

IEEE 7th international conference on Industrial Cyber-Physical systems (ICPS)

作者： Vranjes, Daniel Niggemann, Oliver Helmut Schmidt Univ Inst Automat Technol Hamburg Germany

ISBN: (纸本)9798350363029;9798350363012

In a wide array of engineering disciplines, modular designs serve as a fundamental approach for constructing intricate systems from simpler constituent modules. Domain-specific knowledge about the resulting structures can be formalized, paving the way for informed machine learning, a methodology that leverages prior knowledge to augment data-driven techniques with symbolic insights. One of many possible ways to include prior knowledge into a machine learning pipeline is through incorporation into the design of neural network architectures. We propose a novel algorithm for the learning of modular neural networks on modular cyber-physical systems where structural knowledge is used to build network architectures which resemble technical modules and the information flow is based on semantic relations between data and modules. We investigate potential benefits of modular neural networks compared to classical monolithic approaches based on anomaly detection experiments on data sets of mechanical pendulums with variable numbers of joints which are provided with this research. We define and investigate different levels of knowledge integration and modularity and show that modular designs can have a positive impact on anomaly detection performance. All models and code are published as open source.

关键词： machine learning deep learning informed machine learning neural architecture search modular neural networks cyber-physical systems

来源：评论

学校读者我要写书评

暂无评论

Bayesian Optimization for Sample-Efficient Policy Improvement in Robotic Manipulation

Bayesian Optimization for Sample-Efficient Policy Improvemen...

引用

2024 international conference on Intelligent Robots and systems

作者： Roefer, Adrian Nematollahi, Iman Welschehold, Tim Burgard, Wolfram Valada, Abhinav Univ Freiburg Dept Comp Sci Freiburg Germany Univ Technol Nuremberg Dept Engn Nurnberg Germany

ISBN: (纸本)9798350377712;9798350377705

Sample efficient learning of manipulation skills poses a major challenge in robotics. While recent approaches demonstrate impressive advances in the type of task that can be addressed and the sensing modalities that can be incorporated, they still require large amounts of training data. Especially with regard to learning actions on robots in the real world, this poses a major problem due to the high costs associated with both demonstrations and real-world robot interactions. To address this challenge, we introduce BOpt-GMM, a hybrid approach that combines imitation learning with own experience collection. We first learn a skill model as a dynamical system encoded in a Gaussian Mixture Model from a few demonstrations. We then improve this model with Bayesian optimization building on a small number of autonomous skill executions in a sparse reward setting. We demonstrate the sample efficiency of our approach on multiple complex manipulation skills in both simulations and real-world experiments. Furthermore, we make the code and pre-trained models publicly available at http://***.

关键词： Adversarial machine learning

来源：评论

学校读者我要写书评

暂无评论

Advancing Solar Energy: machine learning Approaches for Predicting Photovoltaic Power Output 18

Advancing Solar Energy: Machine Learning Approaches for Pred...

引用

IEEE 18th international conference on Compatibility, Power Electronics and Power engineering (CPE-POWERENG)

作者： Nassereddine, Kawsar Turzynski, Marek Lukianov, Mykola Strzelecki, Ryszard Gdansk Univ Technol Fac Elect & Control Engn Gdansk Poland

ISBN: (纸本)9798350318272;9798350318265

This research is primarily concentrated on predicting the output of photovoltaic power, an essential field in the study of renewable energy. The paper comprehensively reviews various forecasting methodologies, transitioning from conventional physical and statistical methods to advanced machine learning (ML) techniques. A significant shift has been observed from traditional point forecasting to machine learning-based forecasting in solar power. This transition offers a broader and more detailed perspective for power system operators. The core of this research lies in applying and comparing three distinct machine learning algorithms for forecasting photovoltaic power output. The primary aim is to evaluate each method's accuracy and to identify the algorithm with the lowest prediction error. This comparative analysis is crucial for determining the most effective machine learning forecasting method, significantly contributing to the more reliable and efficient integration of renewable energy into power systems.

关键词： Solar Power Forecasting machine learning Artificial Neural Network Regression Model Lasso's method

来源：评论

学校读者我要写书评

暂无评论

A Comprehensive Analysis of Various Federated learning Techniques 2

A Comprehensive Analysis of Various Federated Learning Techn...

引用

2nd international conference on advances in Computation, Communication and Information Technology, ICAICCIT 2024

作者： Saini, Neha Sharma, Dolly Mehrotra, Deepti Bourazeri, Aikaterini Amity School of Engineering and Technology Amity University Noida India School of Computer Science & Electronic Engineering University of Essex United Kingdom

ISBN: (纸本)9798331541217

A method for data privacy maintenance while training machine learning models across decentralized devices is Federated learning (FL). It comes under the category of decentralized field of machine learning. The decentralized approach faces various challenges, one of them is aggregating model updates which come from diverse and potentially unreliable sources. This study provides a comparison review of various algorithms for privacy preservation, and maintaining security and also talks about the various federated averaging techniques. These techniques are essential for an efficient model aggregation. This research paper also provides an analysis of strengths, weaknesses, and various application areas of federated learning. This paper has included the maximum techniques available for federated learning by comparing their datasets and various other parameters. This paper also provides insight into the related future work that is proposed in various research articles. © 2024 IEEE.

关键词： Federated learning

来源：评论

学校读者我要写书评

暂无评论

Deep learning-based evaluation system for handwritten Chinese character writing standardization 4

Deep learning-based evaluation system for handwritten Chines...

引用

4th international conference on machine learning and Intelligent systems engineering (MLISE)

作者： Yan, Zihao Lai, Sirui Zhang, Ruibin Chen, Liyun Zhuhai Coll Sci & Technol Sch Comp Sci Zhuhai Peoples R China

ISBN: (纸本)9798350375084;9798350375077

In today's society, because of the increase in the frequency of people's use of electronic products, it leads to the phenomenon of irregular writing of Chinese characters and forgetting the characters when putting pen to paper, thus affecting the inheritance of the excellent traditional Chinese culture to a certain extent, so this paper proposes a handwritten Chinese character writing specification evaluation system based on deep learning. This system adopts two deep learning multi-classification models based on ViT to reason about the handwritten Chinese character images in order to obtain the calligraphic evaluation factors, and then calculate the overall evaluation of this handwritten Chinese character through the AHP mathematical evaluation model, and at the same time generate comments for the user through the individual evaluation values.

关键词： evaluation of handwritten Chinese characters deep learning ViT AHP Roboflow

来源：评论

学校读者我要写书评

暂无评论

Sentiment Classification of Product Reviews using machine learning

Sentiment Classification of Product Reviews using Machine Le...

引用

2024 international conference on Recent advances in Electrical, Electronics, Ubiquitous Communication, and Computational Intelligence, RAEEUCCI 2024

作者： Adarsh, Vedullapalli Mohla, Vaibhav Mahto, Ragven Kumar Radhika, P. College of Engineering SRM Institute of Science and Technology Kattankulathur Department of Electronics and Communication Tamilnadu Chengalpattu603203 India

ISBN: (纸本)9798350354539

In the fast-paced e-commerce environment, understanding and learning consumer perceptions of products is essential for businesses to enhance user experience, optimize marketing strategies, and better overall customer satisfaction. This paper investigates the use of machine learning models for sentiment classification and sentiment analysis. VADER, known for its simplicity and speed, employs a sentiment lexicon for polarity scoring. RoBERTa and BERT transformer-based models, which excel in capturing contextual information and long-range dependencies, making them extremely powerful for sentiment analysis. The performance of rules-based models typically falls below machine learning models, which on the other hand, generally exhibit better performance. A diverse dataset is used and the preprocessing steps are detailed. This includes duplicate removal, training of models, using NLP etc. In this paper the highest F1 score is obtained in Bert is 68.9 followed by a 57.8 in RoBERTa and then Vader with a 12.7. These show the superior capability of machine learning models like BERT and RoBERTa in capturing sentiment compared to rule based approaches like Vader which is simpler in nature. This shows the superiority of these models over older models like SVM and Naive Bayes often showing lower average F1 scores. This paper contributes valuable guidance for implementing effective sentiment analysis systems in E-commerce. © 2024 IEEE.

关键词： Sentiment analysis

来源：评论

学校读者我要写书评

暂无评论

Research on Object Detection and Tracking based on Deep learning 4

Research on Object Detection and Tracking based on Deep Lear...

引用

4th international conference on machine learning and Intelligent systems engineering (MLISE)

作者： Pan, Jing Jiangxi Agr Univ Software Coll Nanchang Jiangxi Peoples R China

ISBN: (纸本)9798350375084;9798350375077

With the advancement of artificial intelligence technology, object detection technology in the field of computer vision has played a key role. This article aims to address the accuracy and speed of existing methods in processing live video streams. With the development of deep learning technology, we propose a novel framework of convolutional neural network (CNN) architecture, which optimizes the process of feature extraction and object classification, and significantly improves the detection accuracy. In addition, we have integrated recurrent neural networks (RNNs) to improve the tracking continuity of targets in video sequences. Through the fusion of these technologies, our model not only performs well in multi-target detection, but also reliably tracks targets in the case of occlusion and fast movement. Experimental results show that compared with the existing deep learning methods, the performance of our model on the standard dataset is significantly improved, with a 20% increase in detection speed and a 15% increase in accuracy.

关键词： Objecti detection Convolutional neural network Recurrent neural network Deep learning

来源：评论

学校读者我要写书评

暂无评论

Software Bug Prediction using machine learning 15

Software Bug Prediction using Machine Learning

引用

15th international conference on advances in Computing, Control, and Telecommunication Technologies, ACT 2024

作者： Sathiyaraj, A. Angel, T.G. Ruby Turukmane, Anil V. Ramkumar, D. Guravaiah, Y. Department of Information Technology School of Computing Sathyabama Institute of Science and Technology Tamil Nadu Chennai600 119 India School of Computer Science and Engineering VIT-AP University Inavolu Andhra Pradesh Amaravati522237 India Department of Computer Science and Engineering MJR College of Engineering and Technology Diguvapokulavaripalli Pulicherla Andra Pradesh Chittoor District517214 India

ISBN: (纸本)9798331300579

The entire success of software is impacted by software bug prediction (SBP), a crucial part of the software development and maintenance life cycles. It is necessary to anticipate issues in order to increase the software's dependability, efficiency, and cost. In spite of the fact that a number of methods have been put out in the literature, it is difficult to create a reliable bug prediction model. In this paper, a machine learning (ML) based prediction method for software issues is introduced. Algorithms for guided machine learning that are based on historical data have been used to predict software errors. Two of the classifiers are Support Vector machine and Naive Bayes (NB). How accurate and useful ML methods can be employed was demonstrated through the evaluation process. Included is the utilization of a comparative measure. © Grenze Scientific Society, 2024.

关键词： Prediction models

来源：评论

学校读者我要写书评

暂无评论

Analyzing darknet traffic through machine learning and neucube spiking neural networks

引用

Intelligent and Converged Networks 2024年第4期5卷 265-283页

作者： Iman Akour Mohammad Alauthman Khalid M.O.Nahar Ammar Almomani Brij B.Gupta Information Systems Department College of Computing&InformaticsUniversity of SharjahSharjahUnited Arab Emirates Department of Information Security Faculty of Information TechnologyUniversity of PetraAmman 11196Jordan the Faculty of Computer Studies Arab Open UniversityRiyadh 11681Saudi Arabia Department of Information Technology Al-Huson University CollegeAl-Balqa Applied UniversityIrbid 19117Jordan the Department of Computer Information Science Higher Colleges of TechnologySharjahUnited Arab Emirates the Department of Computer Science and Information Engineering Asia UniversityTaichung 413ChinaSymbiosis Centre for Information TechnologySymbiosis International(Deemed University)Pune 412115India Centre of Inter-disciplinary Research&Innovation University of Petroleum and Energy StudiesDehradun 248007India

The rapidly evolving darknet enables a wide range of cybercrimes through anonymous and untraceable communication *** detection of clandestine darknet traffic is therefore critical yet immensely *** research demonstrates how advanced machine learning and specialized deep learning techniques can significantly enhance darknet traffic analysis to strengthen *** diverse classifiers such as random forest and naïve Bayes with a novel spiking neural network architecture provides a robust foundation for identifying concealed *** on the CIC-Darknet2020 dataset establishes state-of-the-art results with 98%accuracy from the random forest model and 84.31%accuracy from the spiking neural *** pioneering application of artificial intelligence advances the frontiers in analyzing the complex characteristics and behaviours of darknet *** proposed techniques lay the groundwork for improved threat intelligence,real-time monitoring,and resilient cyber defense systems against the evolving landscape of cyber threats.

关键词： darknet traffic machine learning deep learning spiking neural network PCA

来源：评论

学校读者我要写书评

暂无评论

Optimization of Energy Efficiency in IoT inside Smart Grids Utilizing machine learning and Artificial Intelligence Techniques 1

Optimization of Energy Efficiency in IoT inside Smart Grids ...

引用

1st international conference on advances in Computer Science, Electrical, Electronics, and Communication Technologies, CE2CT 2025

作者： Thimmappa, Arunkumar Devalapura Singh, Swati Faculty of Engineering and Technology Department of Mechanical Engineering Karnataka India Maharishi School of Engineering & Technology Maharishi University of Information Technology Uttar Pradesh India

ISBN: (纸本)9798331518578

It is possible for hard jobs to be done by machines with the help of machine learning. Computers and cell phones could make it simpler to regulate the temperature inside a smart grid (a SG), keep an eye on security, and do regular repair. The Internet of Things, or IoT, links the different parts of smart buildings together. As the idea of the Internet of Things (IoT) grows, SGs are being added to bigger networks. The Internet of Things (IoT) is a big part of Smart Grid (SG) s as it makes life better for everyone. It is proven that current methods for life support are harmless & work well to keep people alive. The main purpose of this study is to find out why IoT devices are being put in smart buildings as well as the grid. Looking at it this way, the system that serves IoT devices along with the parts that make them up is very important. Smart grid tracking systems that can be set up remotely can make buildings safer and more comfortable for the people who live or work there. Everything from household technology to SGs needs sensors to work and be watched. Devices that are tied to a network should use less power and can be monitored from afar. Their goal is to help people come up with solutions that use AI, IoT, as well as SGs. The writers also look into SG, networking, and machine intelligence. The last thing we do is look at studies on SG along with IoT. A number of parts of the IoT technology are up for discussion. The first part of this study talks about the most popular ways to use machine learning to predict how much energy a building will need. The authors then talk about IoT and the ways it works, along with SG along with smart meters, and these are needed to get real-time energy information. Then, researcher look into how the different smart grid, Internet of things & machine learning parts work together and combine utilizing basic design with layers that are grouped into entities that talk to each other through links. © 2025 IEEE.

关键词： Smart power grids

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：