检索结果-内蒙古大学图书馆

SSRN 2023年

作者： Qin, Zhaoming Dong, Nanqing Liu, Di Cao, Junwei Automatic Control Laboratory EPFL Lausanne1015 Switzerland Department of Computer Science University of Oxford OxfordOX1 3QD United Kingdom Department of Electrical Engineering Tsinghua University Beijing100084 China Beijing National Research Center for Information Science and Technology Tsinghua University Beijing100084 China

As a data-driven approach, multi-agent reinforcement learning (MARL) has made remarkable advances in solving cooperative residential load scheduling problems. In a simplified scenario, centralized training, the most common paradigm for MARL, precludes large-scale deployment of MARL in communication-constrained cloud-edge environments. As a remedy, distributed training shows unparalleled advantages in real-life application scenarios compared with its centralized counterpart. However, distributed training still faces the challenge of system scalability, e.g., the high communication burden during coordinating individual agents, and needs to comply with data governance in terms of privacy. In this work, we propose a novel MARL solution to address these two practical issues. The proposed solution is based on actor-critic methods, where the global critic is a learned function of individual critics computed solely based on local observations of residences. In this way, the private data of residences is preserved completely and the communication cost is reduced significantly. The simulation experiments demonstrate that the proposed framework can achieve comparable performance to the state-of-the-art actor-critic framework without data governance and communication constraints. © 2023, The Authors. All rights reserved.

关键词： Edge computing

来源：评论

学校读者我要写书评

暂无评论

DeepRicci: Self-supervised Graph Structure-Feature Co-Refinement for Alleviating Over-squashing

DeepRicci: Self-supervised Graph Structure-Feature Co-Refine...

引用

IEEE International Conference on Data Mining (ICDM)

作者： Li Sun Zhenhao Huang Hua Wu Junda Ye Hao Peng Zhengtao Yu Philip S. Yu School of Control and Computer Engineering North China Electric Power University Beijing China School of Computer Science Beijing University of Posts and Telecommunications Beijing China State Key Laboratory of Software Development Environment Beihang University Beijing China Faculty of Information Engineering and Automation Kunming University of Science and Technology Kunming China Department of Computer Science University of Illinois at Chicago IL USA

Graph Neural Networks (GNNs) have shown great power for learning and mining on graphs, and Graph Structure Learning (GSL) plays an important role in boosting GNNs with a refined graph. In the literature, most GSL solutions either primarily focus on structure refinement with task-specific supervision (i.e., node classification), or overlook the inherent weakness of GNNs themselves (e.g., over-squashing), resulting in suboptimal performance despite sophisticated designs. In light of these limitations, we propose to study self-supervised graph structure-feature co-refinement for effectively alleviating the issue of over-squashing in typical GNNs. In this paper, we take a fundamentally different perspective of the Ricci curvature in Riemannian geometry, in which we encounter the challenges of modeling, utilizing and computing Ricci curvature. To tackle these challenges, we present a self-supervised Riemannian model, DeepRicci. Specifically, we introduce a latent Riemannian space of heterogeneous curvatures to model various Ricci curvatures, and propose a gyrovector feature mapping to utilize Ricci curvature for typical GNNs. Thereafter, we refine node features by geometric contrastive learning among different geometric views, and simultaneously refine graph structure by backward Ricci flow based on a novel formulation of differentiable Ricci curvature. Finally, extensive experiments on public datasets show the superiority of DeepRicci, and the connection between backward Ricci flow and over-squashing. Codes of our work are given in https://***/RiemanGraph/.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Preface

Lecture Notes in Electrical Engineering

引用

Lecture Notes in Electrical engineering 2024年 1114 LNEE卷 v页

作者： Lin, Jerry Chun-Wei Shieh, Chin-Shiuh Horng, Mong-Fong Chu, Shu-Chuan Faculty of Automatic Control Electronics and Computer Science Department of Distributed Systems and IT Devices Silesian University of Technology Gliwice Poland

来源：评论

学校读者我要写书评

暂无评论

Gradient Boosting and Deep Learning Models Approach to Forecasting Promotions Efficiency in FMCG Retail 19th

Gradient Boosting and Deep Learning Models Approach to Forec...

引用

19th International Conference on Artificial Intelligence and Soft Computing, ICAISC 2020

作者： Henzel, Joanna Sikora, Marek Department of Computer Networks and System Faculty of Automatic Control Electronics and Computer Science Silesian University of Technology ul. Akademicka 16 Gliwice44-100 Poland

ISBN: (纸本)9783030615338

In the paper, different approaches to the problem of forecasting promotion efficiency are presented. For four defined indicators of promotion effect, prediction models using Gradient Boosting method and Deep Learning methods were trained. The comparison of the results is provided. The experiments were performed for three groups of products from a large grocery company. © 2020, Springer Nature Switzerland AG.

关键词： Forecasting

来源：评论

学校读者我要写书评

暂无评论

An Innovative Smart Irrigation Using Embedded and Regression-Based Machine Learning Technologies for Improving Water Security and Sustainability

引用

IEEE Access 2025年 13卷 100731-100751页

作者： Abdennabi Morchid Abdennacer Elbasri Zahra Oughannou Hassan Qjidaa Rachid El Alami Badre Bossoufi Saleh Mobayen Paweł Skruch LIMAS Laboratory Faculty of Sciences Dhar El Mahraz Sidi Mohamed Ben Abdellah (SMBA) University Fes Morocco Department of Data Science Mundiapolis University of Casablanca Campus Nouaceur Casablanca Morocco Innovation in Mathematics and Intelligent Systems (IMIS) Laboratory Ibn Zohr University Agadir Morocco Systems and Sustainable Environment Laboratory (SED) Faculty of Engineering Sciences (FSI) Private University of Fez (UPF) Fes Morocco Graduate School of Intelligent Data Science National Yunlin University of Science and Technology Douliu Taiwan Energy Systems Research Center Khazar University Baku Azerbaijan Department of Automatic Control and Robotics AGH University of Science and Technology Kraków Poland

Global agriculture is facing major challenges such as food security, sustainable water management, and the preservation of natural resources. Water scarcity, exacerbated by climate change, requires adopting more efficient agricultural practices. Traditional irrigation systems, often imprecise, contribute to water wastage. The use of embedded systems and machine learning offers a solution for optimizing irrigation according to local conditions and actual crop needs while contributing to food security and environmental sustainability. This study proposes an innovative approach to irrigation management, integrating real-time data and predictive models to improve irrigation efficiency. This study proposes an irrigation system based on embedded systems, using sensors and algorithms to collect and analyze data in order to optimize water management. The system adjusts irrigation levels according to specific crop needs, thus contributing to more sustainable water management. Using ML algorithms like linear regression algorithms to model the relationships between environmental factors and crop water requirements, enabling accurate prediction of required irrigation levels based on data collected by sensors. The use of embedded systems such as the ESP32, combined with temperature, humidity, and water level sensors, has enabled the development of an autonomous and efficient system for collecting data in real-time and processing it for decision-making. The proposed model has an MAE of 0.8434, an RMSE of 0.8434, and a coefficient (R2 Score) of 0.4044, offering soil moisture prediction accuracy. Furthermore, the training time of our model is 0.00253 seconds, while the prediction time is 0.00117 seconds. These results show not only the performance of the proposed model in terms of accuracy but also its computational efficiency, outperforming some of the studies mentioned. The results of the study show a significant reduction in water consumption, with a marked improvement in water

关键词： Climate change Embedded systems Smart agriculture Irrigation Food security Optimization methods Machine learning Machine learning algorithms Sustainable development Water resources Water monitoring Environmental management

来源：评论

学校读者我要写书评

暂无评论

2-D GCANet Applied to Denoise 1-D EEG Signals in Online Remote Teaching Scenarios

Journal of Network Intelligence

引用

Journal of Network Intelligence 2023年第4期8卷 1289-1302页

作者： Huang, Ting-Qing Wang, Chuan-Sheng Chen, Zhao-Qi Zhang, Fu-Quan Meng, Xiang-Long Grau, Antoni Chen, Yang Huang, Jing-Wei College of Computer and Control Engineering Minjiang University Fuzhou University Town No. 200 Xiyuangong Road Fuzhou China Department of Automatic Control Technical Polytechnic University of Catalonia Autonomous Region of Catalonia Barcelona Spain College of Computer and Big Data Fuzhou University Fuzhou University Town No. 2 Wulong Jiangbei Avenue Fuzhou China Digital Media Art Key Laboratory of Sichuan Province Sichuan Conservatory of Music No. 2 Wannianchang Street Chenghua District Sichuan Province Chengdu City China Fuzhou Technology Innovation Center of intelligent Manufacturing information System Minjiang University Fuzhou University Town NO. 200 Xiyuangong Road Fuzhou China Fujian Province University No. 599 Quanxiu Road Juyuanzhou Ecological Civilization District Fujian Province Fuzhou City China College of Electronic Engineering Shandong University of Science and Technology No. 579 Qianwangang Road Huangdao District Qingdao China School of Mechanical and Automotive Engineering Fujian University of Technology No. 33 Xuefu South Road University New District Fujian Province Fuzhou City China

When EEG signals are used to assess the level of student engagement in online teaching tasks, they are often interfered by noise. It is a challenge to effectively remove these noises. Currently, deep learning methods have been applied to the field of EEG denoising. However, existing methods still have some problems. The denoised signals still have obvious noise residue, or the original EEG signals are damaged, and the model fitting speed is too slow. Image dehazing, as a typical denoising task in the field of image enhancement, has achieved great success in recent years. Therefore, inspired by advanced models in this field, we introduce CNN into EEG denoising tasks. In this paper, we take GCANet, an excellent image enhancement model, as an example. The dilated convolutions and gate fusion subnetworks included in GCANet enable more efficient EEG signal denoising. The results demonstrate that the proposed model effectively reduces noise while preserving essential features. Furthermore, in comparison to other state-of-the-art models, our proposed model exhibits enhanced robustness and faster convergence, as evidenced by achieving lower loss values after five epochs. Its good performance provides a new development idea for the field of EEG denoising. © 2023, Taiwan Ubiquitous Information CO LTD. All rights reserved.

关键词： Image enhancement

来源：评论

学校读者我要写书评

暂无评论

A Novel Sparse Adaptive Filter for Suppressing Impulsive Disturbance in Audio Signals

SSRN

引用

SSRN 2024年

作者： Zhou, Lei Liu, Hongqing Gan, Lu Zhou, Yi Niedźwiecki, Maciej Truong, Trieu-Kien School of Communications and Information Engineering Chongqing University of Posts and Telecommunications Chongqing400065 China Chongqing Key Lab of Mobile Communications Technology and Intelligent Speech and Audio Research Lab Chongqing University of Posts and Telecommunications Chongqing400065 China College of Engineering Design and Physical Science Brunel University LondonUB8 3PH United Kingdom Department of Automatic Control Faculty of Electronics Telecommunications and Computer Science Gdansk University of Technology Gdansk80-233 Poland Department of Information Engineering I-Shou University 84001 Taiwan

This work studies the sparse adaptive filter designs in the presence of impulsive disturbance for audio signal recovery. By using the sparse representation of desired signal and compressibility of impulsive disturbance, a joint sparse least mean p-norm (JSLMP) optimization, in which p-norm (1 ≤ p ≤ 2) measures the data fidelity and q-norm (0 ≤ q ≤ 1) enforces sparse solutions, is developed, termed as q-JSLMP. The filter weights update is derived based on the gradient descent, and the Adam and variable step size (VSS) are integrated to accelerate convergence and avoid potential local minima. In the special case of q = 1, namely 1-JSLMP, its convergence condition and mean square deviation (MSD) analysis are derived. Finally, the frame of application to corrupted audio signals is designed. The results from experiments demonstrate that the proposed approach outperforms other approaches in terms of objective measures on numerical simulation and speech datasets. © 2024, The Authors. All rights reserved.

关键词： Adaptive filters

来源：评论

学校读者我要写书评

暂无评论

Iteration Complexity of a Fixed-Stepsize SQP Method for Nonconvex Optimization with Convex Constraints 5th

Iteration Complexity of a Fixed-Stepsize SQP Method for Nonc...

引用

5th International Conference on Numerical Analysis and Optimization: Theory, Methods, Applications and Technology Transfer, NAOV 2020

作者： Facchinei, Francisco Kungurtsev, Vyacheslav Lampariello, Lorenzo Scutari, Gesualdo Department of Computer Control and Management Engineering Antonio Ruberti Sapienza University of Rome Rome Italy Department of Computer Science Faculty of Electrical Engineering Czech Technical University in Prague Prague Czech Republic Department of Business Studies Roma Tre University Rome Italy School of Industrial Engineering Purdue University West-LafayetteIN United States

ISBN: (纸本)9783030720391

We consider an SQP method for solving nonconvex optimization problems whose feasible set is convex and with an objective function that is the sum of a smooth nonconvex term and a nonsmooth, convex one. In the proposed method, at each iteration, a direction is generated by solving a strongly convex approximation to the original problem and then a fixed-stepsize is taken in that direction. The complexity result we establish is, as far as we are aware, the best available for the rather general setting we consider. © 2021, The Author(s), under exclusive license to Springer Nature Switzerland AG.

关键词： Fixed-stepsize Iteration complexity SQP methods

来源：评论

学校读者我要写书评

暂无评论

Comparing concepts of quantum and classical neural network models for image classification task

arXiv

引用

arXiv 2021年

作者： Potempa, Rafal Porebski, Sebastian Department of Cybernetics Nanotechnology and Data Processing Faculty of Automatic Control Electronics and Computer Science Silesian University of Technology Gliwice Poland

While quantum architectures are still under development, when available, they will only be able to process quantum data when machine learning algorithms can only process numerical data. Therefore, in the issues of classification or regression, it is necessary to simulate and study quantum systems that will transfer the numerical input data to a quantum form and enable quantum computers to use the available methods of machine learning. This material includes the results of experiments on training and performance of a hybrid quantum-classical neural network developed for the problem of classification of handwritten digits from the MNIST data set. The comparative results of two models: classical and quantum neural networks of a similar number of training parameters, indicate that the quantum network, although its simulation is time-consuming, overcomes the classical network (it has better convergence and achieves higher training and testing accuracy). © 2021, CC BY.

关键词： Image recognition

来源：评论

学校读者我要写书评

暂无评论

Automated Door to Prevent COVID-19 using Fuzzy Logic

Automated Door to Prevent COVID-19 using Fuzzy Logic

引用

International Conference on Electrical, Communication, and computer engineering (ICECCE)

作者： Salah-Ud-Din Khokhar Rana Sohaib Abdul Muneeb Irfana Bibi Muhammad Yasir Noor Muhammad Imran School of Intelligent Manufacturing and Control Engineering Qilu Institute of Technology Jinan P.R.China Department of Computer Science G.C. University Lahore Lahore Pakistan Department of Computer Science Faculty of Computing and Information Technology University of Punjab Lahore Pakistan Department of Electronics G.C. University Lahore Lahore Pakistan

Due to the vagueness and uncertainties due to the coronavirus, it is crucial to implement the standard operating procedures (SOPs) issued by the World Health Organization and the authorities. Therefore an effective control system is required that minimizes the vagueness and uncertainties of people who are on the front line during the period of this pandemic checking the entering persons in an indoor area where it is necessary to maintain a distance of six feet from one another to avoid the spread of coronavirus. In this research work, a robust and efficient system has been proposed that takes protective mask and feverishness values as input and applies them using the fuzzy logic algorithm. If these input values are according to the rules defined in the system, then the door will open as an output. Otherwise, the door will not open. It is, finally, analyzing the output values. This fuzzy model had been implemented using a microcontroller. The genuine difference in the duty cycle of the microcontroller’s output pulse was measured using a 100 MHz digital oscilloscope. Our proposed system was compared with different systems reported in the literature. This work reduces the human effort as well as the negligence of people and builds up the management system without the labor in a mint.

关键词：

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：