检索结果-内蒙古大学图书馆

14th International Conference on Advanced computer Information Technologies, ACIT 2024

作者： Saleh, Eiman M. Imbarake, Tariq Elakeili, Salwa Faraj, Abdul Ghafar Benghazi University Department of Software Engineering Benghazi Libya Benghazi University Department of Computer Science Benghazi Libya Benghazi University Department of Statistics Benghazi Libya

ISBN: (纸本)9798350350036

Nowadays, gas leak detection equipment in the workplace, schools, and homes plays a vital role in protecting public safety. Liquified Petroleum Gas (LPG) leaks are one of the most serious issues in the kitchen because they raise the possibility of an explosion. LPG is the main cooking fuel in Libya. Multiple incidences of household fires, injuries, and deaths have been recorded in Libya as a result of LPG leakage and explosions, as the country's citizens do not have gas leak-detecting systems. This paper aims to present the design and implementation of a smart gas leak detection system based on Arduino. In this system, an Arduino Uno microcontroller with an MQ6 sensor was used to detect gas leakage, an air fan motor to reduce gas concentration by ventilating the air, and GSM as a wireless communication method to notify users via SMS. The proposed system identifies three gas leakage risk levels, each Including its unique scenario. Level one is low-risk, the user is notified Unhealthy status of the dwelling due to gas leakage. level two is medium-risk, the user's friends are informed about the unsafety status due to gas leakage in the user's home. Level three is high-risk;the nearest firefighting organizations are informed of the hazard status. The experimental findings demonstrate that the system is capable of efficiently detecting gas and protecting individuals from a range of undesired situations, as well as being useful in reducing resource and human life losses. © 2024 IEEE.

关键词： Liquefied petroleum gas

来源：评论

学校读者我要写书评

暂无评论

Revisiting Kernel Attention with Correlated Gaussian Process Representation 40

Revisiting Kernel Attention with Correlated Gaussian Process...

引用

40th Conference on Uncertainty in Artificial Intelligence, UAI 2024

作者： Bui, Long Minh Huu, Tho Tran Dinh, Duy Nguyen, Tan Minh Hoang, Trong Nghia FPT Software AI Center Department of Mathematics National University of Singapore Singapore School of Electrical Engineering and Computer Science Washington State University United States

Transformers have increasingly become the de facto method to model sequential data with state-of-the-art performance. Due to its widespread use, being able to estimate and calibrate its modeling uncertainty is important to understand and design robust transformer models. To achieve this, previous works have used Gaussian processes (GPs) to perform uncertainty calibration for the attention units of transformers and attained notable successes. However, such approaches have to confine the transformers to the space of symmetric attention to ensure the necessary symmetric requirement of their GP’s kernel specification, which reduces the representation capacity of the model. To mitigate this restriction, we propose the Correlated Gaussian Process Transformer (CGPT), a new class of transformers whose self-attention units are modeled as cross-covariance between two correlated GPs (CGPs). This allows asymmetries in attention and can enhance the representation capacity of GP-based transformers. We also derive a sparse approximation for CGP to make it scale better. Our empirical studies show that both CGP-based and sparse CGP-based transformers achieve better performance than state-of-the-art GP-based transformers on a variety of benchmark tasks. © 2024 Proceedings of Machine Learning Research. All rights reserved.

关键词： Distribution transformers

来源：评论

学校读者我要写书评

暂无评论

Learning dynamics of kernel-based deep neural networks in manifolds

引用

science China(Information sciences) 2021年第11期64卷 105-119页

作者： Wei WU Xiaoyuan JING Wencai DU Guoliang CHEN School of Computer Science Wuhan University Institute of Deep-sea Science and Engineering Chinese Academy of Sciences School of Computer Guangdong University of Petrochemical Technology Institute of Data Science City University of Macau College of Computer Science and Software Engineering Shenzhen University

Convolutional neural networks(CNNs) obtain promising results via layered kernel convolution and pooling operations, yet the learning dynamics of the kernel remain obscure. We propose a continuous form to describe kernel-based convolutions through integration in neural manifolds. The status of spatial expression is proposed to analyze the stability of kernel-based CNNs. We divide CNN dynamics into the three stages of unstable vibration, collaborative adjusting, and stabilized fluctuation. According to the system control matrix of the kernel, the kernel-based CNN training proceeds via the unstable and stable status and is verified by numerical experiments.

关键词： learning dynamics kernel-based convolution manifolds control model network stability

来源：评论

学校读者我要写书评

暂无评论

Real-time 3D hand-object pose estimation for human-robot interaction

Real-time 3D hand-object pose estimation for human-robot int...

引用

2022 Australasian Conference on Robotics and Automation, ACRA 2022

作者： Yin, Yue McCarthy, Chris Rezazadegan, Dana Department of Computer Science and Software Engineering Swinburne University of Technology Melbourne Australia

ISBN: (纸本)9781713866480

Fast and accurate real-time 3D hand-object pose estimation is an important capability for many collaborative human-robot interaction applications. However, the application of vision-based hand-object pose estimation remains challenging, with most existing approaches relying on dedicated depth-sensing cameras and/or specialised hardware support to meet high computation and memory storage requirements. This limits the generality of use-cases for which 3D hand-object pose estimation models can be applied. Recent work has considered the use of lightweight models to reduce storage and computational overheads, typically trading off accuracy for realtime performance. While sufficient for some applications, high accuracy remains a key priority, especially in the context of human-robot collaboration. Addressing this issue, we propose a novel lightweight model for 3D handobject pose estimation that maintains realtime performance while also achieving an accuracy exceeding that of current state-of-theart approaches on the First Person Hand Action (FPHA) dataset. Our model thus offers a viable RGB-only solution for human-robot collaboration applications, and for low compute, embedded platforms more generally. © 2022 Australasian Robotics and Automation Association. All rights reserved.

关键词： Human robot interaction

来源：评论

学校读者我要写书评

暂无评论

Anomaly-based network intrusion detection using Transfer learning 2

Anomaly-based network intrusion detection using Transfer lea...

引用

2nd International Conference on Multidisciplinary engineering and Applied science, ICMEAS 2023

作者： Ola-Obaado, Similoluwa Suleiman, Muhammad Aliyu Nile University of Nigeria Computer Science Department Abuja Nigeria Nile University of Nigeria Software Engineering Department Abuja Nigeria

ISBN: (纸本)9798350358834

The rapid development of computer networks and network applications, along with the present global increase in hacking and computer network attacks, has increased the demand for stronger intrusion detection and prevention solutions. The intrusion detection system (IDS) is crucial in detecting irregularities and attacks on the network, which has expanded in size and pervasiveness. In this paper, we develop an intrusion detection system based on deep learning and leverage a recent security dataset, CSE-CIC-IDS2018, to implement a more realistic IDS. We proposed an approach for network intrusion detection using Transfer learning to address both the binary and multiclass classification of network attacks. Network features from the dataset are transformed into grayscale images, the images are used as input to the ResNet50 deep learning model. The dataset is imbalanced, therefore, to increase the generalization of model, the imbalance ratio is reduced by using a synthetic data generation model called Synthetic Minority Oversampling Technique (SMOTE). Experimental results show that our model achieves good performance for both binary and multiclass classification with accuracy of 92 percent for each case. The proposed approach provides a new method for intrusion detection using transfer learning and SMOTE for correcting data imbalance. © 2023 IEEE.

关键词： Intrusion detection

来源：评论

学校读者我要写书评

暂无评论

A Comparison of Deep Learning and Machine Learning Methods for Water Quality Prediction 3

A Comparison of Deep Learning and Machine Learning Methods f...

引用

3rd International Conference on Intelligent Data Communication Technologies and Internet of Things, IDCIoT 2025

作者： Srinivasa Gupta, N. Valarmathi, B. Hiwanj, Ishaan Department of Manufacturing Engineering School of Mechanical Engineering Vellore Institute of Technology Tamil Nadu Vellore India Department of Software and Systems Engineering School of Computer Science Engineering and Information Systems Vellore Institute of Technology Tamil Nadu Vellore India Department of Information Technology School of Computer Science Engineering and Information Systems Vellore Institute of Technology Tamil Nadu Vellore India

ISBN: (纸本)9798331527549

Predicting water quality is essential to preserving human health and environmental sustainability. Traditional water quality assessment methods often face scalability and real-time monitoring limitations. With accuracies of 62%, 72%, 83%, 69%, 63%, 66%, 71%, 63%, and 64%, respectively, the current techniques utilized were Logistic Regression, Decision Trees, Random Forest Regressor, Extreme Gradient Boosting, Naive Bayes, K-nearest neighbors, Support Vector Machine, AdaBoost, and Bagging [9]. This study addresses these challenges by leveraging Adaptive Synthetic Sampling (ADASYN) to balance the dataset and evaluating model performance on datasets of 5,000 and 10,000 entries per class. A robust dataset obtained from Kaggle was used, with five models – Long Short-Term Memory (LSTM), Feed Forward Neural Network (FFNN), Categorical Boosting (CatBoost), Extreme Gradient Boosting (XGBoost), and Random Forest – evaluated and compared. The proposed methods demonstrate significant improvements in accuracy, with XGBoost achieving the highest accuracy of 95.53%, followed by Random Forest at 93.98%. This work underscores the importance of advanced machine learning techniques in addressing the limitations of traditional methods, enhancing accuracy, scalability, and adaptability in water quality prediction. These findings contribute to advancing environmental monitoring and management practices with reliable, data-driven insights. © 2025 IEEE.

关键词： Support vector machines

来源：评论

学校读者我要写书评

暂无评论

Using Deep Reinforcement Learning and Formal Verification in Safety Critical Systems: Strategies and Challenges 23

Using Deep Reinforcement Learning and Formal Verification in...

引用

23rd IEEE International Conference on software Quality, Reliability, and Security Companion, QRS-C 2023

作者： Sharma, Satyam Rahim, Muhammad Abdul Basit Ur Hussain, Shahid Abid, Muhammad Rizwan Liu, Tairan California State University Computer Engineering and Computer Science Long BeachCA United States Computer Science and Software Engineering Penn State University Behrend PA United States Florida Polytechnic University Computer Science LakelandFL United States Mechanical and Aerospace Engineering California State University Long BeachCA United States

ISBN: (纸本)9798350359398

Deep Reinforcement Learning (DRL) is critical in modern Artificial Intelligence (AI), powering innovations from gaming to autonomous vehicles. As DRL continues its rapid ascent, ensuring its systems are both trustworthy and effective is crucial. This research focuses on different DRL techniques and the challenges faced in real-life scenarios. The paper also describes various formal verification techniques and the challenges related to their application. It sheds light on the different frameworks and tools that can enhance the credibility of systems. We performed an extensive literature survey to present the existing methodologies, tools, and frameworks. The analysis systematically reviews and categorizes various formal verification techniques and frameworks employed in DRL. The insights garnered from this study are anticipated to foster an enriched understanding of the processes and contribute to decision-making in Safety Critical Systems using DRL and verification. © 2023 IEEE.

关键词： Decision making

来源：评论

学校读者我要写书评

暂无评论

Knowledge-Assisted DRL for Energy Harvesting Based Multi-Access Wireless Communications 23

Knowledge-Assisted DRL for Energy Harvesting Based Multi-Acc...

引用

23rd IEEE International Conference on High Performance Computing and Communications, 7th IEEE International Conference on Data science and Systems, 19th IEEE International Conference on Smart City and 7th IEEE International Conference on Dependability in Sensor, Cloud and Big Data Systems and Applications, HPCC-DSS-SmartCity-DependSys 2021

作者： Luo, Sheng Zhang, Haodi Li, Qifan Wu, Kaishun College of Computer Science and Software Engineering Shenzhen University Shenzhen518060 China

ISBN: (纸本)9781665494571

This paper investigates the use of deep reinforcement learning (DRL) to maximize the long-term throughput of multi-access wireless communication systems. Specifically, we consider the scenario in which an access point (AP) without dedicated power supply harvests energy from the ambient environment and uses the harvested energy to deliver data packets to multiple users. We formulate the access control problem as a Markov Decision Process (MDP) with unknown system dynamics and apply the deep reinforcement learning (DRL) approach to solve this problem, i.e., the optimal access strategy is approximated by using a double deep Q-network (DDQN). As the considered system usually has large state and action spaces, some domain knowledge is embedded into the training process to improve the performance of DDQN. It is shown that with the help of domain knowledge, access control strategies which can achieve a higher system throughput can be found by the knowledge-embedded DDQN. Moreover, the performance of the access control policy obtained by using the knowledge-embedded DDQN is compared with that of several conventional transmission schemes. Experiment results show that the transmission policy obtained by using our proposed model can achieve better performance. © 2021 IEEE.

关键词： Access control Wireless communication Training System dynamics Smart cities Power supplies Process control

来源：评论

学校读者我要写书评

暂无评论

Small object detection based on WavesNET 10

Small object detection based on WavesNET

引用

10th International Conference on Information Systems and Computing Technology, ISCTech 2022

作者： Li, Zhengnan Zhao, Ji Lu, Xiaowei School of Computer and Software Engineering University of Science and Technology Liaoning Anshan China

ISBN: (纸本)9798350332933

In recent years, object detection technology has developed rapidly, especially the FPN has been widely adopted in production application and academic research. Compared with the traditional network, the accuracy of EfficientDet based on the FPN structure has reached 55.1% on the COCO dataset, and the number of parameters and computation cost is also greatly reduced by 9-25 times. However, the EfficientDet training process does not integrate the upper and lower level feature information quickly enough, especially depending on the multi-layer BiFPN overlap and increasing input size and channel number, therefore, we proposed WavesNet based on EfficientDet. First, we integrated seven channels on the basis of BiFPN to make all levels of the fusion more quickly and effectively, and then applied the ComC (Combine Convolution) extraction method in the initial phase of feature extraction. In addition, the EfficientDet fast normalization formula was improved to ensure the numerical stability of the features. Experimental results show the WavesNet has 30% faster convergence rate and 0.5% higher accuracy on COCO dataset than the EfficientDet. The detection effect on small targets is obvious. © 2022 IEEE.

关键词： Object recognition

来源：评论

学校读者我要写书评

暂无评论

Towards Secure Agile software Development Process: A Practice-Based Model 48

Towards Secure Agile Software Development Process: A Practic...

引用

48th Euromicro Conference on software engineering and Advanced Applications, SEAA 2022

作者： Ardo, Abdulhamid A. Bass, Julian M. Gaber, Tarek University of Salford Department of Computer Science & Software Engineering Manchester United Kingdom

ISBN: (纸本)9781665461528

Agile methods are a well-established paradigm in the software development field. Agile adoption has contributed to improving software quality. However, software products are vulnerable to security challenges and susceptible to cyberattacks. This study aims to improve security of software products when using an agile software development process. A multi-methods qualitative research approach was adopted in this study. First, we conducted semi-structured interviews with 23 agile practitioners having varied years of cybersecurity experiences. An approach informed by grounded theory methodology was adopted for data analysis. Second, we developed a novel practice-based agile software development process model derived from the results of the data analysis conducted. Third, we validated the model through a focus group comprising five senior agile cybersecurity professionals to evaluate its relevancy and novelty. The study has identified 26 security practices, organized into the six - software development life-cycle phases: planning, requirements, design, implementation, testing, and deployment. We have mapped the practices onto four swim lanes each representing an agile role. The self-organizing team is exclusively involved in three security practices, the security specialist in nine, penetration tester in one and the DevOps team collaborates on one with the security specialist. There are also seven practices that are collaboratively performed by the self-organizing team and the security specialist. Each of the practices in the model was examined during the validation phase of the study. There are two contributions in this study. First, the paper proposes a novel practice-based model comprising of 26 security practices mapped to agile roles. Second, we propose a new practice, in response to an observed lack of collaborative ceremonies, to disseminate awareness of and hence compliance with security standards. © 2022 IEEE.

关键词： software testing

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：