检索结果-内蒙古大学图书馆

Decomposition of Differences of Independent and Identically Distributed Random Variables

Moscow University Computational Mathematics and Cybernetics 2025年第1期49卷 67-72页

作者： Ushakov, V. G. Ushakov, N. G. Department of Computational Mathematics and Cybernetics Moscow State University Moscow Russia Institute of Informatics Problems Federal Research Center ‘‘Computer Science and Control’’ Russian Academy of Sciences Moscow Russia Norwegian University of Science and Technology Trondheim Norway

The authors consider conditions where the distribution of the components of the difference between two independent identically distributed random variables can be uniquely reconstructed with an accuracy of up to a shift and reflection. This uniqueness is essential for solving a number of characterization problems in mathematical statistics. An algorithm for estimating the components is presented for when data are given in a symmetrized form.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Koopman Form of Nonlinear Systems with Inputs

arXiv

引用

arXiv 2022年

作者： Iacob, Lucian Cristian Tóth, Roland Schoukens, Maarten Control Systems Group Eindhoven University of Technology Eindhoven Netherlands Systems and Control Lab Institute for Computer Science and Control Budapest Hungary

The Koopman framework proposes a linear representation of finite-dimensional nonlinear systems through a generally infinite-dimensional globally linear embedding. Originally, the Koopman formalism has been derived for autonomous systems. In applications for systems with inputs, generally a linear time invariant (LTI) form of the Koopman model is assumed, as it facilitates the use of control techniques such as linear quadratic regulation and model predictive control. However, it can be easily shown that this assumption is insufficient to capture the dynamics of the underlying nonlinear system. Proper theoretical extension for actuated continuous-time systems with a linear or a control-affine input has been worked out only recently, however extensions to discrete-time systems and general continuous-time systems have not been developed yet. In the present paper, we systematically investigate and analytically derive lifted forms under inputs for a rather wide class of nonlinear systems in both continuous and discrete time. We prove that the resulting lifted representations give Koopman models where the state transition is linear, but the input matrix becomes state-dependent (state and input-dependent in the discrete-time case), giving rise to a specially structured linear parameter-varying (LPV) description of the underlying system. We also provide error bounds on how much the dependency of the input matrix contributes to the resulting representation and how well the system behaviour can be approximated by an LTI Koopman representation. The introduced theoretical insight greatly helps for performing proper model structure selection in system identification with Koopman models as well as making a proper choice for LTI or LPV techniques for the control of nonlinear systems through the Koopman approach. © 2022, CC BY-NC-SA.

关键词： Continuous time systems

来源：评论

学校读者我要写书评

暂无评论

Hierarchical Clustering-Based State Grouping Reinforcement Learning for Switching Decision of Autonomous Vehicles

Hierarchical Clustering-Based State Grouping Reinforcement L...

引用

2023 IEEE International Conference on Unmanned Systems, ICUS 2023

作者： Ouyang, Wenjie Jiao, Yiwen Liu, Yang Li, Yang Hu, Manjiang Qin, Hongmao College of Mechanical and Vehicle Engineering Hunan University Changsha China College of Computer Science and Electronic Engineering Hunan University Changsha China College of Electrical Mechanical and Computer Science Delft University of Technology Delft Netherlands College of Mechanical and Vehicle Engineering Hunan University State Key Laboratory of Advanced Design and Manufacturing Technology for Vehicle Changsha China Institute of Hunan University Wuxi Intelligent Control Research Wuxi China

ISBN: (纸本)9798350316308

Reinforcement learning (RL) has gained wide attention, but its implementation in autonomous vehicles is still limited by insufficient sample efficiency and heavy training costs. The training efficiency of RL agents is influenced by the dimension of the state space, which can be partitioned to reduce the complexity of sampling and computation. This study proposes a hierarchical clustering-based state grouping reinforcement learning (HCSG-RL) method for the switching decision of autonomous vehicles. First, we partition the base state space into groups and generate a hierarchical tree of state space groups. Then, we train multiple sub-agents for each node in the hierarchical tree. Finally, we add these trained-well sub-model into master policy. This method allows us to fully explore all state spaces and improve the training efficiency of individual agents, which handles the 'long-tail' issue and the curse of dimensionality issue. We conduct experiments in a simulation environment and results show that the proposed method has 16-72% reward improvement compared to the tree model in different road length. © 2023 IEEE.

关键词： Reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

Impact of diameter on D-shaped POF sensors for salt detection

Impact of diameter on D-shaped POF sensors for salt detectio...

引用

International Laser technology and Optics Symposium 2024, iLATOS2024

作者： Khan, Abdul Ali Thirunavakkarasu, Punithavathi M. bte Saidin, Norazlina Noor, Ahmad Shukri Muhammad Waqas, Nawaf Communication Technology Section Universiti Kuala Lumpur British Malaysian Institute Kuala lumpur53100 Malaysia Faculty of I.C.T BUITEMS Quetta87100 Pakistan Department of Electrical and Computer Engineering International Islamic University Malaysia Kuala Lumpur53100 Malaysia Department of Computer and Communication Systems Engineering Faculty of Engineering Universiti Putra Malaysia Selangor Malaysia Department of Instrumentation and Control Engineering Universiti Kuala Lumpur British Malaysian Institute Bt. 8 Jalan Sungai Pusu Selangor Gombak53100 Malaysia

This study investigates the relationship between the diameter of a D-shaped plastic optical fiber (POF) sensor and its optical response as a function of varying salt concentrations. A simple and efficient dry etching polishing technique was employed to create D-shaped POF sensors with a range of diameters. All the fabricated sensors were carefully developed to maintain a consistent length of 20 cm with a sensing region of 1cm. To evaluate the optical intensity response, each sensor, with its respective diameter, was examined with the salt concentration ranging from (0 to 5%) examination utilizing a spectrometer. The sensor response was characterized by quantifying the variations observed in the intensity of the optical signal propagating through the sensing region. The sensors having a diameter of 500 μm, 700 μm, and 900 μm had a sensitivity of 149 A.U/%salt, 66 A.U/%salt, and 11.2 A.U/%salt respectively. The sensitivity of the 500 μm diameter sensor increased 13 times approximately as compared to the sensor having diameter of 900 μm. © 2025 institute of Physics Publishing. All rights reserved.

关键词： Dry etching

来源：评论

学校读者我要写书评

暂无评论

Convolutional Neural Network System for Melanoma Identification

Convolutional Neural Network System for Melanoma Identificat...

引用

International Conference on control, Automation, Robotics and Vision (ICARCV)

作者： Radu Marian Macovei Loretta Ichim Dan Popescu Faculty of Automatic Control and Computer Science National University of Science and Technology POLITEHNICA Bucharest Bucharest Romania “Stefan S. Nicolau” Institute of Virology Bucharest Romania

ISBN: (数字)9798331518493

ISBN: (纸本)9798331518509

Melanoma is a very dangerous type of skin cancer and its detection in the early stages is necessary for proper treatment. The article proposes an intelligent system, based on the fusion of the decisions of several neural networks to increase the performance in melanoma detection from dermatoscopic images. The system implementation method is based on the optimal choice of the number and type of neural networks involved by testing the possible combinations. According to the selection procedure, a system was implemented with four neural networks DenseNet 201, VGG 19, MobileNet, and EfficientNet. The results obtained on two different databases (ISIC 2019 - for learning, validation, and testing - and PH2 for testing) were better than those obtained on individual networks or in other works from literature.

关键词： Automation Databases Melanoma Performance analysis Convolutional neural networks Intelligent systems System implementation Robots Testing

来源：评论

学校读者我要写书评

暂无评论

Research on Grayscale Images Flame Detections Based on Colorazion Algorithm and YOLOv8

Research on Grayscale Images Flame Detections Based on Color...

引用

IEEE International Workshop on Imaging Systems and Techniques (IST)

作者： Peng Liu Zhenfu Sun Biao Zhou Hongyu Wang Haozhen Qu Institute of Energy and Electrical Engineering Changchun University of Science and Technology Changchun China School of Control and Computer Engineering North China Electric Power University Beijing China

ISBN: (数字)9798350378214

ISBN: (纸本)9798350378221

Flame detection algorithms are crucial for real-time fire monitoring using surveillance cameras. Current flame detection algorithms perform excellently on color cameras; however, many night vision cameras can only capture grayscale images, making flame detection less effective in such scenarios. Traditional flame detection algorithms rely heavily on color features, especially the red and yellow hues of flames. Conversely, grayscale images lack color information, making these algorithms ineffective. To address this limitation, we propose a novel flame detection algorithm that integrates the DeOldify model with the you only look once(YOLOv8) model. DeOldify can effectively add color to grayscale images. YOLOv8 is a powerful object detection model that excels in identifying various objects within an image. This algorithm leverages DeOldify's colorization capabilities and YOLOv8's powerful object detection abilities to enhance flame detection accuracy in grayscale images. This algorithm has been tested multiple times on public datasets, converting original images to grayscale and then colorizing them to construct the dataset. The results show that, compared to grayscale images, this algorithm improves the mean average precision (mAP@0.5) by 19.7% on different validation sets, by 4% on different training sets, and by 12.9% on grayscale images captured by night vision cameras. By comparing the training results of models such as YOLOv8n and YOLOv8s, the colorized images improve the model's performance. This demonstrates the high generalizability of this algorithm.

关键词： Night vision YOLO Training Image color analysis Surveillance Fires Gray-scale Cameras Real-time systems Detection algorithms

来源：评论

学校读者我要写书评

暂无评论

Optimal Synthesis of LTI Koopman Models for Nonlinear Systems with Inputs

arXiv

引用

arXiv 2022年

作者： Iacob, Lucian Cristian Tóth, Roland Schoukens, Maarten Control Systems Group Eindhoven University of Technology Eindhoven Netherlands Systems and Control Laboratory Institute for Computer Science and Control Budapest Hungary

A popular technique used to obtain linear representations of nonlinear systems is the so-called Koopman approach, where the nonlinear dynamics are lifted to a (possibly infinite dimensional) linear space through nonlinear functions called observables. In the lifted space, the dynamics are linear and represented by a so-called Koopman operator. While the Koopman theory was originally introduced for autonomous systems, it has been widely used to derive linear time-invariant (LTI) models for nonlinear systems with inputs through various approximation schemes such as the extended dynamics mode decomposition (EDMD). However, recent extensions of the Koopman theory show that the lifting process for such systems results in a linear parameter-varying (LPV) model instead of an LTI form. As LTI Koopman model based control has been successfully used in practice and it is generally temping to use such LTI descriptions of nonlinear systems, due to the simplicity of the associated control tool chain, a systematic approach is needed to synthesise optimal LTI approximations of LPV Koopman models compared to the ad-hoc schemes such as EDMD, which is based on least-squares regression. In this work, we introduce optimal LTI Koopman approximations of exact Koopman models of nonlinear systems with inputs by using 2-gain and generalized H2 norm performance measures. We demonstrate the advantages of the proposed Koopman modelling procedure compared to EDMD. © 2022, CC BY-NC-SA.

关键词： Nonlinear systems

来源：评论

学校读者我要写书评

暂无评论

Training the DNN of a Single Observer by Conducting Individualized Subjective Experiments

Training the DNN of a Single Observer by Conducting Individu...

引用

International Workshop on Quality of Multimedia Experience, QoMEx

作者： Pavel Majerl Lohic Fotio Tiotsop Marcus Barkowsky Deggendorf Institute of Technology University of Applied Sciences Deggendorf Germany Control and Computer Engineering Department Politecnico di Torino Torino Italy

Predicting the quality perception of an individual subject instead of the mean opinion score is a new and very promising research direction. Deep Neural Networks (DNNs) are suitable for such prediction but the training process is particularly data demanding due to the noisy nature of individual opinion scores. We propose a human-in-the-loop training process using multiple cycles of a human voting, DNN training, and inference procedure. Thus, opinion scores on individualized sets of images were progressively collected from each observer to refine the performance of their DNN. The results of computational experiments demonstrate the effectiveness of our approach. For future research and benchmarking, five DNNs trained to mimic five observers are released together with a dataset containing the 1500 opinion scores progressively gathered from each of these observers during our training cycles.

关键词：

来源：评论

学校读者我要写书评

暂无评论

CONCENTRATION IN GOSSIP OPINION DYNAMICS OVER RANDOM GRAPHS

arXiv

引用

arXiv 2023年

作者： Xing, Yu Johansson, Karl H. Division of Decision and Control Systems School of Electrical Engineering and Computer Science KTH Royal Institute of Technology and Digital Futures Stockholm Sweden

We study concentration inequalities in gossip opinion dynamics over random graphs. In the model, a network is generated from a random graph model with independent edges, and agents interact pairwise randomly over the network. During the process, regular agents average neighbors’ opinions and then update, whereas stubborn agents do not change opinions. To approximate the original process, we introduce a gossip model over an expected graph, obtained by averaging all possible networks generated from the random graph model. Using concentration inequalities, we derive high-probability bounds for the distance between the expected final opinion vectors over the random graph and over the expected graph. Leveraging matrix perturbation results, we show how such concentration can help study the effect of network structure on the expected final opinions in two cases: (i) When the influence of stubborn agents is large, the expected final opinions polarize and are close to stubborn agents’ opinions. (ii) When the influence of stubborn agents is small, the expected final opinions are close to each other. With the help of concentration inequalities for Markov chains, we obtain high-probability bounds for the distance between time-averaged opinions and the expected final opinions over the expected graph. In simulation, we validate the theoretical findings, and study a gossip model over a stochastic block model that has community structure. © 2023, CC BY.

关键词： Dynamics

来源：评论

学校读者我要写书评

暂无评论

Efficient Tool Condition Monitoring Using SOM-Hebb Classifier on FPGA

Efficient Tool Condition Monitoring Using SOM-Hebb Classifie...

引用

International Conference on Contemporary Computing and Informatics (IC3I)

作者： A. Somani Akanksha Kokate R. D. Mishra Abhilasha Department of Electronics & Computer Engineering Maharashtra Institute of Technology Aurangabad India Department of Instrumentation and Control Engineering College of Engineering Pune India

Tool condition monitoring is vital for enhancing productivity, reducing costs, and improving product quality in manufacturing industries. Existing approaches, such as threshold-based methods, sensor-based methods, and machine learning-based techniques, have limitations in terms of accuracy, adaptability, interpretability, and computational costs. This research proposes a tool condition monitoring system based on the Self-Organizing Map (SOM) algorithm and the Hebbian learning algorithm. The system is implemented on a Field-Programmable Gate Array (FPGA) platform for real-time processing in manufacturing environments. A dataset is acquired using sensors, pre-processed, and divided into training and testing sets. The SOM-Hebb classifier is designed with appropriate parameters, and the tool condition monitoring algorithm involves projection histograms, discrete Fourier transform (DFT), and the SOM-Hebb classifier. The FPGA implementation achieves high-speed processing and meets timing constraints. Simulation results demonstrate the system's functionality and accuracy, with a classification accuracy of 99.22%. The system exhibits a processing time of 35 clock cycles. This research provides a promising approach for efficient and accurate tool condition monitoring in manufacturing industries.

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：