检索结果-内蒙古大学图书馆

Proceedings of the 41st International Conference on Machine Learning

作者： Gugan Thoppe Prashanth L. A. Sanjay P. Bhat Dept. of Computer Science and Automation Indian Institute of Science (IISc) Bengaluru India and Robert Bosch Centre for Data Science and Artificial Intelligence IIT Madras Chennai India Dept. of Computer Science and Engineering Indian Institute of Technology Madras Chennai India TCS Research Hyderabad India

We tackle the problem of estimating risk measures of the infinite-horizon discounted cost of a Markov cost process. The risk measures we study include variance, Value-at-Risk (VaR), and Conditional Value-at-Risk (CVaR). First, we show that estimating any of these risk measures with ε-accuracy, either in expected or high-probability sense, requires at least Ω(1/ε2) samples. Then, using a truncation scheme, we derive an upper bound for the CVaR and variance estimation. This bound matches our lower bound up to logarithmic factors. Finally, we discuss an extension of our estimation scheme that covers more general risk measures satisfying a certain continuity criterion, such as spectral risk measures and utility-based shortfall risk. To the best of our knowledge, our work is the first to provide lower and upper bounds for estimating any risk measure beyond the mean within a Markovian setting. Our lower bounds also extend to the infinite-horizon discounted costs' mean. Even in that case, our lower bound of Ω(1/ε2) improves upon the existing Ω(1/ε) bound (Metelli et al., 2023).

关键词：

来源：评论

学校读者我要写书评

暂无评论

Efficient 3D Transformer with cluster-based Domain-Adversarial Learning for 3D Medical Image Segmentation

Efficient 3D Transformer with cluster-based Domain-Adversari...

引用

IEEE International Symposium on Biomedical Imaging

作者： Haoran Zhang Hao Chen Sch. of Artificial Intelligence and Automation Huazhong University of Science and Technology Dept. of Computer Science and Engineering The Hong Kong University of Science and Technology Dept. of Chemical and Biological Engineering The Hong Kong University of Science and Technology

Real-world application of volumetric medical image segmentation is still challenging due to the domain shift problem and the disability to process volumetric information efficiently by existing algorithms. To address these problems, we propose a 3D Swin Transformer with a pyramidal downsampling strategy to process volumetric information efficiently, dubbed as PDSwin. Specifically, the improved 3D Swin Transformer includes a spatial downsampling strategy that downsamples 2D slices pyramidally according to the spatial relationship, reducing the computation complexity while providing a wider downsampled receptive field. Furthermore, we propose a cluster-based domain-adversarial learning algorithm to attenuate the domain shift problem. The algorithm generates fine-grained cluster-based domains instead of employing center-based domains, ameliorating the domain-adversarial learning performance. We evaluated our model against other competitive models on brain stroke lesion segmentation and prostate segmentation tasks. Extensive experimental results indicated that our proposed model outperforms other models, demonstrating the efficacy of our proposed method.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Ontological Analysis of the Digital Crime Conceptual Model

Ontological Analysis of the Digital Crime Conceptual Model

引用

International computer Sciences and Information Technologies (CSIT)

作者： Lidiia Vlasenko Nataliia Lutska Nataliia Zaiets Vitaliy Lysenko dept. of Software Engineering and Cyber Security State University of Trade and Economics Kyiv Ukraine dept. of Automation and Computer Technologies of Control Systems National University of Food Technologies Kyiv Ukraine dept. of Automation and Robotic Systems National University of Life and Environmental Sciences of Ukraine Kyiv Ukraine

A conceptual ontological model of Digital Crime has been developed, consisting of five non-empty classes. Identification and classification were carried out according to the experience of domestic and foreign experts. The central place of the model is occupied by objects of the Digital Crime class, which have their own attributes and are connected by various links with other objects. Objects of some classes are filled in by users, others are formed by knowledge engineers. Based on the ontological analysis, SWRL-rules are constructed that automatically form links between objects of the custom classes Digital crime and Digital evidence with objects of the Type of crime and Criminal Liability classes. The last class contains articles of the Criminal Code of Ukraine on criminal liability for digital crimes. The ontological model is developed taking into account possible requests to it. Such queries can be formed on the basis of direct relationships between human and automatic concepts. And also contain logical chains to concepts that do not contain an explicit link in the ontology. It is assumed that the model can be expanded and supplemented. The ontological model of Digital Crime can be used to share information, develop new systems and applications, improve information retrieval and ensure its accordance.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Solution for Agri Disaster Using Climate Changes Adaptation by Ml 6

Solution for Agri Disaster Using Climate Changes Adaptation ...

引用

6th International Conference on Contemporary Computing and Informatics, IC3I 2023

作者： Ganesh, L. Venkatesh, R. Prasanna Kumar, Kaushal Kumari, Khushboo Borkakoty, Sangeeta Raju, A. Rajalakshmi Engineering College Robotics and Automation Dept. Tamilnadu Chennai India SACS MAVMM Engineering College Department of Mechanical Engineering Tamilnadu Madurai India School of Life Sciences Department of Botany Mahatma Gandhi Central University Bihar India University of Science & Technology Department of Computer Science & Electronic Meghalaya India KLEF Vaddeswaram Green Field Department of Commerce Vijayawada India

ISBN: (纸本)9798350304480

This paper proposes using good sized getting to know fashions to recognize lesions of cotton leaves in the context of pictures of the reaping within the subject. Cotton is among the maximum economically essential plant harvests, as it's far grown all over the international. Its spread in tropical areas has made it the goal of a wide variety of farming irritations and illnesses, necessitating succesful plans. moreover, the consequences of commonplace disturbances and disorders are difficult to isolate inside the early stages, and the author may additionally battle to locate the suitable visible verification of a scientific difficulty. To help in resolving the difficulty, it contemporary inquiry proposes a solution that takes into consideration substantial training within the filtering of cotton leaves, making it possible to reveal the cotton yield's health and make higher judgments approximately its management. The studying models GoogleNet like Resnet50, which use convolutional cerebrum associations, performed an accuracy of 86.6 percentage and 89.2 percentage, respectively. assist vector machines, Closest k-pals, faux creativeness associations, and neuro-tender are some of the extraordinary as well as everyday tactics again for treatment of photographs. The convolutional cerebellum networks ended up being 25% extra unique, implying that this method can upload to a more strong and short investigation of the flora looking to fill inside the field. © 2023 IEEE.

关键词： Artificial Intelligence Convolutional Neural Networks Image Processing Precision Agriculture

来源：评论

学校读者我要写书评

暂无评论

Robot Trajectron: Trajectory Prediction-based Shared Control for Robot Manipulation

Robot Trajectron: Trajectory Prediction-based Shared Control...

引用

IEEE International Conference on Robotics and automation (ICRA)

作者： Pinhao Song Pengteng Li Erwin Aertbeliën Renaud Detry Dept. Mechanical Engineering Research Unit Robotics Automation and Mechatronics KU Leuven Leuven Belgium College of Computer Science and Software Engineering Shenzhen University Shenzhen China Flanders Make@KU Leuven Dept. Electrical Engineering Research Unit Processing Speech and Images KU Leuven Leuven Belgium

ISBN: (数字)9798350384574

ISBN: (纸本)9798350384581

We address the problem of (a) predicting the trajectory of an arm reaching motion, based on a few seconds of the motion’s onset, and (b) leveraging this predictor to facilitate shared-control manipulation tasks, by reducing the operator’s cognitive load through assistance in their anticipated direction of motion. Our novel intent estimator, dubbed the Robot Trajectron (RT), produces a probabilistic representation of the robot’s anticipated trajectory based on its recent position, velocity and acceleration history. By taking arm dynamics into account, RT can capture the operator’s intent better than other SOTA models that only use the arm’s position, making it particularly well-suited to assist in tasks where the operator’s intent is susceptible to change. We derive a novel shared-control solution that combines RT’s predictive capacity to a representation of the locations of potential reaching targets. Our experiments demonstrate RT’s effectiveness in both intent estimation and shared-control tasks. We will make the code and data supporting our experiments publicly available at https://***/detry-lab/public/robot-trajectron

关键词： Codes Dynamics Estimation Grasping Probabilistic logic Cognitive load Trajectory

来源：评论

学校读者我要写书评

暂无评论

An Improved YOLOv5 Detection Method for Strip Surface Defect

An Improved YOLOv5 Detection Method for Strip Surface Defect

引用

International Conference on automation and Computing (ICAC)

作者： Yefeng Liu Longfei Yu Qichun Zhang dept. Liaoning Key Laboratory of Information Physical Fusion and Intelligent CNC Machine Tool Manufacturing Shenyang Institute of Technology dept. School of Automation and Electrical Fushun China dept. School of Automation and Electrica Engineering Shenyang Ligong University dept. Department of Computer Science University of Bradford Bradford U. K.

Traditional steel surface defect detection methods have issues such as low detection accuracy and efficiency. It is suggested a technique based on enhanced YOLOv5 steel surface fault identification. In order to improve the network's capacity to handle feature information, this study suggests adding SE attention mechanisms to the YOLOv5s network's Backbone, Neck, and SPPF and replacing the C3 structure at the end of the Backbone with C3TR. To achieve network model lightweight, it is proposed to replace the Neck part in the YOLOv5s model and the convolution in SPPF with Ghost convolution. Then, to improve network feature fusion, extend the improved SPPF to the Neck part. Introduce the SPD convolution in the Backbone section to reduce the information loss caused by the neural network's convolution. The tested mAP is 75.3%, which is 5.3% higher than the original model, 24.6% higher than the SSD method, and 4.6% higher than the YOLOv6 algorithm after the upgraded model has been trained on the NEU-DET dataset. The observed FPS is 81.3, which is 6.3 more than the relative literature algorithm and 18.0 more than the SSD algorithm.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Low-Latency and Energy-Efficient Task Scheduling for End-Edge-Cloud Collaborative Computing

Low-Latency and Energy-Efficient Task Scheduling for End-Edg...

引用

IEEE International Conference on Systems, Man and Cybernetics

作者： Jingyao Li Haitao Yuan Yaofei Ma Jing Bi Jinhong Yang Jia Zhang School of Automation Science and Electrical Engineering Beihang University Beijing China College of Computer Beijing University of Technology Beijing China CSSC Systems Engineering Research Institute Beijing China Dept. of Computer Science Southern Methodist University Dallas TX USA

ISBN: (数字)9781665410205

ISBN: (纸本)9781665410212

Mobile edge computing (MEC) is a new paradigm that improves the quality of service compared with traditional cloud computing. In MEC, computational tasks are submitted by numerous end users and are partially offloaded to edge servers or a central cloud. However, the characteristics of tasks are different from each other, and the limited resources of computational nodes are also heterogeneous, which brings great challenges to computation offloading and resource allocation for MEC. This work establishes an end-edge-cloud collaborative computing network, which consists of end devices, edge servers, and a central cloud. Task execution location and CPU running frequency determine the execution time and energy consumption to finish the tasks. Considering the aforementioned factors, a multi-objective constrained optimization problem is formulated. To solve the problem, an improved Non-dominated Sorting Genetic Algorithm II (NSGA-II) with self-adaptive crossover and mutation rates is proposed, which is called Improved NSGA-II with _Self-adaptive Crossover and Mutation (INSCM). The total execution time and energy consumption can be jointly minimized with our proposed INSCM. Numerous experiments are carried out to test the performance of INSCM. Simulation results show that INSCM effectively improves the performance of NSGA-II and surpasses random offloading and NSGA-III, which shows practical use in real-life scenarios.

关键词： Energy consumption Cloud computing Time-frequency analysis Multi-access edge computing Simulation Collaboration Servers Optimization Sorting Genetic algorithms

来源：评论

学校读者我要写书评

暂无评论

Flow-rate Feeder Control Design With PI-PI Cascade Controller on Biomass Power Plant

Flow-rate Feeder Control Design With PI-PI Cascade Controlle...

引用

International Conference on Advanced Mechatronics, Intelligent Manufacture, and Industrial automation (ICAMIMIA)

作者： Indrawan Budi Prasetyo Aris Budiyarto Ciptian Weried P Imam Wahyudi Farid Setya Widyawan Prakosa Jenq-Shiou Leu Dept. Electronic and Computer Engineering National Taiwan University of Science and Technology Taipei Taiwan PT. Parametrik Solusi Integrasi Bandung Indonesia Dept. Automation Electrical Engineering Institut Teknologi Sepuluh Nopember Surabaya Indonesia

The goal of gasification-based biomass power plants in this research is to be able to produce 5–10 KW of electricity through the thermal energy conversion process. Syngas, a byproduct of the burning of biomass, is created during the gasification process and may be utilized to power generators. The gasification process will diminish biomass, therefore if it reaches a particular level, it has to be frequently supplied to the feeder. Uncontrolled feeder transport capacity will result in the biomass that feeds the gasifier exceeding or falling short of the need, causing fluctuations in the amount of syngas generated. Due to this, the PI-PI cascade controller was designed to regulate biomass transport capacity. From the design of the PI-PI Cascade control system, the parameters for the primary loop controller Kp = 38.9; Ki = 0.81, and the secondary loop Kp = 3.5; Ki = 5. The results of the control design obtained time constant (ti) 0.0101 seconds; rise time(tr) for tr 10% at 0.014 seconds and tr 90% at 0.674 seconds; settling time(ts) of the response signal with Ts 0.05% at 9.5 seconds; delay time (td) at 0.065 seconds and if a disturbance is given when the condition is steady, it will return to the initial condition at 3.21 seconds after the disturbance is given.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Is Almost Sure Asymptotic Stabilizability Sufficient for Input Delay Tolerability of Nonlinear Stochastic Systems?

引用

IEEE Transactions on Automatic Control 2025年

作者： Yu, Xin Lin, Wei Jiangsu Normal University School of Electrical Engineering and Automation Xuzhou221116 China Case Western Reserve University Dept. of Electrical Computer and Systems Engineering ClevelandOH44106 United States

This article considers the problem of input delay tolerance (IDT) for stochastic control systems that are almost surely asymptotically stabilizable but inherently nonlinear (e.g., neither locally Lipschitz continuous nor linear growth and non-smoothly stabilizable). Using the notions of almost sure forward completeness [42] and weighted homogeneity, together with the existence of the solutions of continuous time-delay stochastic systems, we prove under Lyapunov-like conditions that : i) almost sure asymptotic stabilizability by nonsmooth homogeneous state feedback and dominated homogeneity of degree zero imply input delay tolerability of stochastic nonlinear systems, and ii) almost sure asymptotic stabilizability by homogeneous output feedback also implies input delay tolerability of the same class of multi-input/multi-output stochastic nonlinear systems. Our results contain not only the existing IDT results for the global Lipschitz or linear growth stochastic systems as special cases, but also important applications to significant classes of stochastic control systems with unstabilizable and undetectable linearization. © 1963-2012 IEEE.

关键词： Stochastic control systems

来源：评论

学校读者我要写书评

暂无评论

Control of Microgrid Cluster Using Raspberry Pi

Control of Microgrid Cluster Using Raspberry Pi

引用

Energy Technologies for Future Grids (ETFG), IEEE International Conference on

作者： David Carrasco-González Raúl Sarrias-Mena Pablo Horrillo-Quintero Francisco Llorens-Iborra Luis M. Fernández-Ramírez Department of Electrical Engineering SURET Research Group University of Cadiz (UCA) Spain Dept. of Engineering in Automation Electronics and Computer Architecture and Networks SURET Reseach Group UCA Spain

Currently, microgrid clusters (MGC) are being increasingly used owing to their benefits and human needs. MGs can be composed of power generation, consumption, and storage elements. They permit the combination of DC and AC technologies, even for a single microgrid (MG). In this paper, a new control system of a MGC consisting of two MGs connected to the main grid is developed. One of them is a DC MG that is composed of a wind turbine (WT), electrolyzer (EZ), ultracapacitor (UC), fuel cell (FC) and DC loads. The other MG is an AC MG, which is made up of a PV generator, a battery bank and AC loads. The MGC is tested under different working conditions in an experimental setup composed of a Raspberry Pi microcontroller, an OPAL-RT 4510 unit and a PC host. The experimental setup allows the system to run in real time and validate the results obtained. Finally, the control system implemented in the MGC exhibits a correct behaviour under different conditions, according to the results obtained.

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：