检索结果-内蒙古大学图书馆

10th International conference on control, Instrumentation and Automation, ICCIA 2024

作者： Farooghi, Kian Khaloozadeh, Hamid K. N. Toosi University of Technology Department of Systems and Control Tehran Iran

ISBN: (纸本)9798331509972

This paper presents a novel multi-Agent deep reinforcement learning (DRL) approach for automated trading in the foreign exchange (Forex) market, utilizing Proximal Policy Optimization (PPO). The proposed system employs two cooperative agents: one specializing in position entry and another in position exit. Both agents leverage Long Short-Term Memory (LSTM) neural networks to process temporal market data and technical indicators. The reward function is designed to encourage collaborative behavior, with rewards shared between agents upon trade closure. Furthermore, the implementation of Generalized Advantage Estimation (GAE) method enables the agents to balance immediate rewards with long-Term consequences, enhancing the stability and efficiency of the learning process. This research contributes to the growing field of AI-driven Forex trading strategies, offering valuable insights into applying multi-Agent reinforcement learning in dynamic financial markets. The proposed methodology shows efficiency in developing lucrative trading systems capable of navigating the complexities of the Forex market. Experiments conducted on EUR/USD data demonstrate the effectiveness of this approach. Across 150 runs on the test dataset, the framework achieved profits ranging from 3.5% to 6.9%, with a median return of 5.8%. © 2024 ieee.

关键词： Deep reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

data-driven Decision Making and Near-Optimal Path Planning for Multiagent System in Games

IEEE JOURNAL ON MINIATURIZATION FOR AIR AND SPACE SYSTEMS

引用

ieee JOURNAL ON MINIATURIZATION FOR AIR AND SPACE systems 2023年第3期4卷 320-328页

作者： Wang, Xindi Liu, Hao Gao, Qing Beihang Univ Sch Astronaut Beijing 100191 Peoples R China Beihang Univ Inst Artificial Intelligence Beijing 100191 Peoples R China Zhongguancun Lab Beijing 100094 Peoples R China Beihang Univ Sch Automat Sci & Elect Engn Beijing 100191 Peoples R China

In this article, the optimal real-time decision making and near-optimal path planning problem for multiagent systems subject to bounded state, collision avoidance, external disturbance, and partially unknown nonlinear dynamics of the multiagent system in complex games, is addressed and applied to the unmanned aerial vehicle. A mean-field decision-making model based on the neighbor information is established to transform the decision-making problem into a Bellman equation solving problem. A data-driven dynamic programming algorithm is proposed to solve the Bellman equation and generate an optimal strategy using the data from the historical database and expert knowledge. The near-optimal path planning problem is formulated with an optimal coordination control problem, and an online integral reinforcement learning algorithm is proposed to iteratively interact with the environment to obtain a near-optimal path. Simulation results are provided to verify the effectiveness of the proposed methods.

关键词： Decision making Path planning Heuristic algorithms Reinforcement learning databases Games Collision avoidance multiagent system path planning reinforcement learning unmanned aerial vehicle (UAV)

来源：评论

学校读者我要写书评

暂无评论

Distribution Network Digital Twin: a Basic Machine learning based Voltage Estimator 14

Distribution Network Digital Twin: a Basic Machine Learning ...

引用

14th ieee International conference on Communications, control, and Computing Technologies for Smart Grids, SmartGridComm 2023

作者： Iaali, Mohamad El Razi, Reza Bruyere, Antoine Francois, Bruno Soares, Joao Univ. Lille Arts et Metiers Institute of Technology Centrale Lille Junia Ulr 2697 - L2EP LilleF-59000 France Gecad - Res. Group on Intelligent Engineering and Computing for Advanced Innovation and Development Lasi - Intelligent Systems Associate LAboratory Polytechnic of Porto Porto Portugal

ISBN: (纸本)9781665455541

The increasing integration of renewable energy sources, and the widespread use of electric vehicles have introduced greater complexity and dynamics in power flows in distribution systems. With bidirectional power flows enhancing the observability of the grid becomes essential to identify critical feeders. In this context, data-driven methods offer a promising approach to enhance situational awareness for power system operators and improve their understanding of the system behavior under variabilities. This paper proposes a voltage estimation method based on a Machine learning algorithm to address this challenge. By leveraging data-driven techniques, the proposed method aims to provide accurate and efficient voltage estimations in distribution networks, contributing to the effective management and control of modern power systems. © 2023 ieee.

关键词： Distributed power generation

来源：评论

学校读者我要写书评

暂无评论

Computational Cost Analysis and data-driven Predictive Modeling of Cloud-Based Online-NILM Algorithm

引用

ieee TRANSACTIONS ON CLOUD COMPUTING 2022年第4期10卷 2409-2423页

作者： Asres, Mulugeta Weldezgina Ardito, Luca Patti, Edoardo Politecn Torino Dept Control & Comp Engn I-10129 Turin Italy

Online non-intrusive load monitoring algorithms have captivated academia and industries as parsimonious solutions for household energy efficiency monitoring as well as a safety control, anomaly detection, and demand-side management. However, the computational energy cost for executing such algorithms should not overcome the promised energy efficiency from the disaggregated appliance specific consumption information feed-backs. Moreover, the energy efficiency of cloud computing systems is also becoming a concern for the environment due to carbon emission. This study analyzes the energy spent to execute NILM algorithms via computation cost estimation and prediction using computing system-level power monitoring and data-driven approaches. A generic framework for an automated algorithm cost monitoring and modeling methodologies is devised for large load scale deployment of Cloud-based Online-NILM algorithms. The efficacy of the proposed approach was examined and validated on two computing system use-cases, i.e., Dedicated Server and Cloud Virtual Server. The prediction models, developed using statistical and machine learning tools, demonstrate the promising applicability of the data-driven approach with a very high prediction accuracy without detailed knowledge of the computing systems and the algorithm.

关键词： NILM load disaggregation data-driven computational cost algorithm machine learning cloud systems

来源：评论

学校读者我要写书评

暂无评论

Harnessing Deep learning for Fault Detection in Industry 4.0: A Multimodal Approach 6

Harnessing Deep Learning for Fault Detection in Industry 4.0...

引用

6th International conference on Cognitive Machine Intelligence

作者： Shen, Jialie Morrison, Marie Miao, Haiyan Gu, Fengshou City St Georgias Univ London Dept Comp Sci London England Univ Bristol Sch Engn Math & Technol Bristol Avon England Univ Huddersfield Sch Comp & Engn Huddersfield W Yorkshire England

ISBN: (纸本)9798350386738;9798350386721

The advent of Industry 4.0 has reshaped the modern industries (e.g., Manufacturing, Automotive, Aerospace and Defense), driven by the rapid development of artificial intelligence, smart sensing technologies, and interconnected cyber-physical systems. One of the most important goal for Industry 4.0 is the improvement and enhancement in productivity, reduction of production losses, and operational efficiency. High quality fault detection within different environments is the foundation to achieve these goals. Industry 4.0 emphasizes the use of interconnected systems, smart sensors, and advanced analytics, which generate large volumes of data from production environments. The data complexity and scale, product by the widespread deployment of sensors and IoT devices, demands sophisticated Machine learning (ML) and data-driven methods. In this paper, we argue that the future of accurate and reliable fault detection lies in the seamless fusion of edge-cloud computing, explainable AI, and adaptive machine learning algorithms capable of processing high-frequency, multimodal data streams in real time. Different kinds of advantages for multimondal learning and core enabling techniques are also discussed and revise. This vision paper outlines the future trajectory of multimodal learning based fault detection, via identifying critical research gaps and promising directions for future study.

关键词： Deep learning Fault Detection Multimodal learning Industry 4.0

来源：评论

学校读者我要写书评

暂无评论

Optimization of ReLU Activation Function for Deep-learning-based Breast Cancer Classification on Mammogram Images

Optimization of ReLU Activation Function for Deep-Learning-b...

引用

ieee International conference on Automatic control and Intelligent systems (I2CACIS)

作者： Razali, Noor Fadzilah Isa, Iza Sazanita Sulaiman, Siti Noraini Osman, Muhammad Khusairi Karim, Noor Khairiah A. Nordin, Siti Aminah Univ Teknol MARA Coll Engn Elect Engn Studies Permatang Pauh Campus George Town Malaysia Univ Sains Malaysia Bertam Adv Med & Dent Inst Dept Biomed Imaging George Town Malaysia Univ Teknol MARA Coll Engn Elect Engn Studies Pasir Gudang Campus Johor Baharu Malaysia

ISBN: (纸本)9798350372113;9798350372106

The use of the deep Convolutional Neural Network (CNN) in breast cancer classification of mammogram images has been widely investigated to aid radiologists in better clinical diagnoses. Multiple levels of convolution and non-linearity repetitions in CNN's architecture are required to extract significant data to be represented. However, the vanishing gradient effect occurs when deeper network training as a product of the partial derivative of loss function on each weightage update can cause no meaningful network learning, even with additional epochs. Overcoming this using the activation function of rectified linear unit (ReLU) by allowing neurons to be activated to allow non-linearity when the output is more than zero could lessen the problem. However, restrictive allowance of non-linearity for <0 for final feature extraction when producing output probability on highly complex data such as mammogram images leads to dropped network performance. To overcome this, this study proposed an adaptive ReLU based on genetic algorithm (GA) profiling to determine the best threshold value for allowing neuron activation based on mutation and adaptation to improve the restrictive capability of the original ReLU. We modified the adaptive ReLU on the final learning layer of two CNN architectures and observed the performance on a public mammogram dataset of INbreast. Our experiments show improved accuracy from 95.0% to 98.5% and improved classification performance compared to other well-known activation functions. Applying evolutionary-based GAs to activation functions can represent an exciting frontier in meta-learning for neural networks.

关键词： activation function ReLU Convolutional neural network mammogram genetic algorithm optimization

来源：评论

学校读者我要写书评

暂无评论

PMSM System Identification by Knowledge-informed Machine learning 22

PMSM System Identification by Knowledge-informed Machine Lea...

引用

22nd ieee International conference on Industrial Informatics, INDIN 2024

作者： Yu, Yue Tong, Jiageng Xia, Jinhui Su, Jinya Li, Shihua School of Automation Southeast University Nanjing China Key Laboratory of Measurement and Control of Complex Systems of Engineering Ministry of Education Southeast University Nanjing China

ISBN: (纸本)9798331527471

Because of its excellent efficiency, compact dimensions, and accurate control features, Permanent Magnet Synchronous Motor (PMSM) are experiencing widespread applications across various industries. By accurately characterizing the dynamic behavior of PMSM systems through system identification, engineers can ensure that PMSM motors reach their maximum potentials and meet the stringent requirements of modern industrial and technical systems while reducing energy consumption and maintenance costs. The traditional recursive least squares method is sensitive to noises, and unable to accurately identify parameters in complex environments. Pure data-driven models lack interpretability and require complex model architecture and computational costs. To this end, this work draws knowledge-informed neural ordinary differential equations (NODEs) for system identification, which embeds system prior knowledge into the NODEs for more efficient and accurate model learning. Comparative simulations show that this method not only obtains a higher-precision system model, but also significantly reduces the amount of training data and computation costs. © 2024 ieee.

关键词： Cost reduction

来源：评论

学校读者我要写书评

暂无评论

PhysORD: A Neuro-Symbolic Approach for Physics-infused Motion Prediction in Off-road Driving

PhysORD: A Neuro-Symbolic Approach for Physics-infused Motio...

引用

2024 International conference on Intelligent Robots and systems

作者： Zhao, Zhipeng Li, Bowen Du, Yi Fu, Taimeng Wang, Chen Univ Buffalo Inst Artificial Intelligence & Data Sci Dept Comp Sci & Engn Spatial AI & Robot SAIR Lab Buffalo NY 14260 USA Carnegie Mellon Univ Robot Inst Pittsburgh PA 15213 USA

ISBN: (纸本)9798350377712;9798350377705

Motion prediction is critical for autonomous off-road driving, however, it presents significantly more challenges than on-road driving because of the complex interaction between the vehicle and the terrain. Traditional physics-based approaches encounter difficulties in accurately modeling dynamic systems and external disturbance. In contrast, data-driven neural networks require extensive datasets and struggle with explicitly capturing the fundamental physical laws, which can easily lead to poor generalization. By merging the advantages of both methods, neuro-symbolic approaches present a promising direction. These methods embed physical laws into neural models, potentially significantly improving generalization capabilities. However, no prior works were evaluated in real-world settings for off-road driving. To bridge this gap, we present PhysORD, a neural-symbolic approach integrating the conservation law, i.e., the Euler-Lagrange equation, into data-driven neural models for motion prediction in off-road driving. Our experiments showed that PhysORD can accurately predict vehicle motion and tolerate external disturbance by modeling uncertainties. It outperforms existing methods both in accuracy and efficiency and demonstrates data-efficient learning and generalization ability in long-term prediction.

关键词： Off road vehicles

来源：评论

学校读者我要写书评

暂无评论

Trust-based positional forgery detection in AI-driven autonomous intersection management using hybrid graph-based reinforcement learning 24

Trust-based positional forgery detection in AI-driven autono...

引用

24th ieee International conference on data Mining Workshops, ICDMW 2024

作者： Anand, Shajina Seton Hall University Department of Mathematics and Computer Science United States

ISBN: (纸本)9798331530631

Positional forgery by malicious vehicles can severely disrupt AI-driven autonomous intersection management systems, leading to inaccurate traffic predictions and increased accident risks. Traditional detection methods struggle in multi-tiered vehicular networks. This paper presents a Trust-Based Positional Forgery Detection (TPFD) system that leverages a Hybrid Graph-based Reinforcement learning (HGRL) framework to combat positional forgery. The system integrates Graph Neural Networks (GNNs) for modelling Vehicle-to-Everything (V2X) communication, Recurrent Neural Networks (RNNs) for analysing temporal data, and Reinforcement learning (RL) for dynamic trust evaluation. TPFD achieved a detection accuracy of 98.3%, reduced the false positive rate to 3.5%, and scaled efficiently to networks of up to 10,000 vehicles. The system also reduced hazardous traffic gap errors by 30%, enhancing intersection safety. This research strengthens the reliability and security of AI-based traffic systems, contributing to more efficient traffic management and increased public trust in autonomous vehicles. © 2024 ieee.

关键词： Recurrent neural networks

来源：评论

学校读者我要写书评

暂无评论

A data-driven Machine learning Model for the Stress-Strain Behavior of Single Grain SAC305 Solder Joints 74

A Data-Driven Machine Learning Model for the Stress-Strain B...

引用

ieee 74th Electronic Components and Technology conference (ECTC)

作者： Mondal, Debabrata Suhling, Jeffrey C. Mirkoohi, Elham Lall, Pradeep Auburn Univ Dept Mech Engn Auburn AL 36849 USA Auburn Univ Ctr Adv Vehicle & Extreme Environm Elect Auburn AL 36849 USA

ISBN: (纸本)9798350375985;9798350375992

Sn-Ag-Cu (SAC) alloys are a type of lead-free solder that are commonly used in microelectronic packaging. These materials typically contain more than 95% beta-Sn, which is a highly anisotropic material with different elastic moduli and coefficients of thermal expansion depending on the crystal direction. In particular, a set of 6 unique elastic compliances are needed to model the elastic behavior. When beta-Sn is plastically deformed beyond its elastic limit, the deformations are governed by the movement of dislocations on specific crystallographic planes and directions. There are 10 different slip families with 32 slip systems that have been identified for beta-Sn body-centered tetragonal (BCT) crystals. In our current work, we are developing approaches to modeling the mechanical response of SAC solder joints made up of a small number of beta-Sn grains reinforced with Ag3Sn and Cu6Sn5 intermetallic particles. Anisotropic elasticity theory for a BCT crystal is being used to model the elastic behavior of the grains, while Crystal Plasticity (CP) theory is being used to predict the post yielding material response by considering the crystallographic deformation mechanisms. In the study reported in this paper, a data-driven machine learning (ML) model was developed to predict the mesoscale deformation behavior of single crystal SAC samples. The ML model was trained using a large number of solder stress-strain curves for various crystal orientations obtained using simulations performed with the crystal plasticity-based DAMASK code. The constants in the CP code were first calibrated with experimental solder stress-strain data (both elastic and plastic regions) to obtain the model constants that describe the overall material deformation of SAC305 alloy from a crystallographic standpoint. For a single crystal SAC sample, there are an infinite number of possible orientations for uniaxial loading that can lead to an infinite number of different overall mechanical deformation

关键词： Crystal Plasticity Beta-Tin Slip systems BCT Dislocation Machine learning data driven Time Series LSTM

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：