检索结果-内蒙古大学图书馆

2025 ieee International conference on Acoustics, Speech, and Signal Processing, ICASSP 2025

作者： Cederberg, Daniel Larsson, Erik G. Johansson, Mikael Department of Electrical Engineering Stanford University Stanford United States Department of Electrical Engineering Linköping University Linköping Sweden Division of Decision and Control Systems KTH Royal Institute of Technology Stockholm Sweden

ISBN: (纸本)9798350368741

This paper presents a comprehensive approach to federated learning in wireless networks. We discuss communication strategies that address packet loss and bitrate limitations in both uplink and downlink transmissions, and introduce FABLE, a novel optimization algorithm designed to operate effectively under these network constraints. The algorithm also supports non-smooth regularizers and accommodates heterogeneous data distributions across clients. We provide theoretical convergence guarantees under gradient compression and asynchronous operation, and demonstrate the efficiency of our approach through numerical experiments. © 2025 ieee.

关键词： Federated learning optimization algorithms wireless communications

来源：评论

学校读者我要写书评

暂无评论

Reinforcement learning for Sustainable Combustion in Thermal Power Plant 1

Reinforcement Learning for Sustainable Combustion in Thermal...

引用

1st ieee International conference on Smart and Sustainable Developments in Electrical Engineering, SSDEE 2025

作者： Azad, S.M.A.K. Siraparapu, Sri Ramya Sri Venkateswara College of Engineering Department of Electronics and Communication Engineering Andhra Pradesh Tirupati India School of Electronics Engineering VIT-AP University Andhra Pradesh Amaravati India

ISBN: (纸本)9798331542108

In recent years, the optimization of thermal power plants has become a critical area of research, driven by the urgent need to enhance energy efficiency and reduce emissions. This paper proposes a novel approach leveraging Reinforcement learning (RL) techniques to optimize the operation of thermal power plants in real-time. RL, a branch of machine learning, offers a promising framework for decision-making in dynamic and uncertain environments, making it well-suited for the complex and nonlinear nature of power plant operations. By integrating RL algorithms with streaming data from sensors and control systems, our proposed methodology enables adaptive and autonomous control of key parameters such as air-fuel ratio and combustion temperature. This approach not only maximizes combustion efficiency but also minimizes emissions of harmful pollutants such as NOx, SOx, and particulate matter. Through simulation studies and real-world experiments, we demonstrate the effectiveness of our proposed approach, paving the way for the next generation of intelligent and sustainable power plant operation. We present a conceptual framework for RL-based control of thermal power plants highlighting potential benefits such as combustion efficiency, improved environmental performance, and enhanced operational flexibility. © 2025 ieee.

关键词： Reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

Tuning Variable Speed Wind Turbine controller Using a data-driven Approach: The TD3 Reinforcement learning Algorithm 12

Tuning Variable Speed Wind Turbine Controller Using a Data-D...

引用

12th Iranian conference on Renewable Energies and Distributed Generation, ICREDG 2025

作者： Avarzaman, Hossein Ahmadi Manzar, Mojtaba Nouri Badri, Vahid Shahid Beheshti University Faculty of Electrical Engineering Tehran Iran

ISBN: (纸本)9798331525132

The Proportional Integral Derivative (PID) controller and its variants are widely employed in industrial systems due to their well-documented benefits. This paper investigates the implementation of reinforcement learning (RL) algorithms for tuning controller parameters in the pitch control of a variable-speed wind turbine operating in the full-load region. Due to the complex dynamical behavior of wind turbines, tuning the controller has significant challenges. The proposed method is model-free and conducts controller tuning solely based on system's data. The Twin Delayed Deep Deterministic Policy Gradients (TD3) algorithm, an off-policy Actor-Critic method, is employed for tuning the parameters as it does not necessitate a system model. The efficacy of the TD3- based tuning method is demonstrated through simulations on 4.8 MW variable speed variable pitch wind turbine, highlighting the effectiveness of the proposed methodology. © 2025 ieee.

关键词： Two term control systems

来源：评论

学校读者我要写书评

暂无评论

Automated Inter-Hospital Communication for Bed Allocation 4

Automated Inter-Hospital Communication for Bed Allocation

引用

4th International conference on Sentiment Analysis and Deep learning, ICSADL 2025

作者： Akila Priya Varshini, M. Kashvi, R. Pushpa Varshini, A. Nirmala, V. Vidhya, V. Dept of Artificial Intelligence and Data Science Chennai India

ISBN: (纸本)9798331523923

This paper presents an automated inter-hospital communication system for efficient bed allocation during pandemics, featuring role-based access for secure data handling. The system leverages real-time data integration, predictive analytics, and AI-driven optimization to streamline bed allocation across hospitals. The role-based access makes sure that only authorized personnel can access information, enhancing data security and operational transparency. This approach addresses common inefficiencies in current systems, reducing patient wait times and improving resource management in high-demand scenarios, ultimately enhancing crisis response capabilities. © 2025 ieee.

关键词： data integration

来源：评论

学校读者我要写书评

暂无评论

RD-OpenMax: Rethinking OpenMax for Robust Realistic Open-Set Recognition

引用

ieee TRANSACTIONS ON NEURAL NETWORKS AND learning systems 2025年第4期36卷 7565-7579页

作者： Yin, Xiaojie Cao, Bing Hu, Qinghua Wang, Qilong Tianjin Univ Coll Intelligence & Comp Tianjin 300350 Peoples R China Haihe Lab Informat Technol Applicat Innovat Tianjin 300350 Peoples R China

Open-set recognition (OSR) toward a practical open-world setting has attracted increasing research attention in recent years. However, existing OSR settings are either too idealized or focus on specific scenes such as long-tailed distribution and few-shot samples, which fail to capture the complexity of real-world scenarios. In this article, we propose a realistic OSR (ROSR) setting that covers a diverse range of challenging and real-world scenarios, including fine-grained cases with strong semantic correlation and a large number of species, few-shot samples, long-tailed sample distribution, dynamic inputs (e.g., images, spatio-temporal, and multimodal signals) and cross-domain adaptation. In particular, we rethink the simple and basic OpenMax for the ROSR setting and introduce a novel method, regularized discriminative OpenMax (RD-OpenMax), to handle the challenges in the ROSR setting. RD-OpenMax improves upon the basic OpenMax approach by introducing a covariance attention-based covariance pooling (CACP) module as a global aggregation step before the deep architecture's classifier. This module explores rich statistical information on features and provides discriminative distance scores for OpenMax. To address the instability of extreme value theory (EVT) estimation due to insufficient training samples under few-shot and long-tailed scenarios, we propose a regularized EVT (REVT) method based on Monte Carlo sampling to recalibrate the distribution of distance scores. As such, our RD-OpenMax performs a REVT model of distance scores generated by discriminative CACP representations to distinguish known classes and recognize unknown ones effectively and robustly. Extensive experiments are conducted on more than ten visual benchmarks across several scenarios, and the empirical comparisons show that the ROSR setting challenges existing state-of-the-art OSR approaches. Moreover, our RD-OpenMax clearly outperforms its counterparts under the ROSR setting while performing fav

关键词： Semantics Correlation Training Benchmark testing Estimation Birds Weibull distribution Covariance attention-based covariance pooling (CACP) OpenMax realistic open-set recognition (ROSR) regularized extreme value theory (EVT)

来源：评论

学校读者我要写书评

暂无评论

Intelligent Defenses: Advancing Cybersecurity through Machine learning-driven Malware Detection 14

Intelligent Defenses: Advancing Cybersecurity through Machin...

引用

14th ieee International conference on Communication systems and Network Technologies, CSNT 2025

作者： Mittal, Sakshi Rajvanshi, Prateek United States

ISBN: (纸本)9798331531935

The exponential rise in cyberthreats has rendered malware detection a crucial component of cybersecurity. Traditional signature-based detection technologies are insufficient for combating emerging threat vectors, as increasingly complex malware targets sensitive systems. Machine learning techniques have emerged as a viable alternative for the precise detection and classification of malware through data-driven insights. This study presents a comprehensive malware detection system that combines Convolutional Neural Networks (CNN) with sophisticated machine learning (ML) methodologies, including ensemble techniques like Random Forest and Gradient Boosted Decision Trees. To rectify class imbalance and enhance model performance, the framework utilizes extensive preprocessing methods, including SMOTE. The Random Forest achieved the highest accuracy of 97.4%, followed by Gradient Boosted Decision Trees at 96.7%;the CNN demonstrated competitive performance at 95.0%. Dynamic API call sequences were identified as essential elements for precise classification. The proposed technique significantly reduces false negatives, hence ensuring improved protection against malicious software and enhancing recall. This research provides a scalable framework for future malware detection systems and demonstrates how machine learning may enhance cybersecurity protocols. © 2025 ieee.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

Exploring IoT and Machine learning Integration for Soil Nutrients Monitoring and Crop Recommendation: A Survey 3

Exploring IoT and Machine Learning Integration for Soil Nutr...

引用

3rd International conference on Electronics and Renewable systems, ICEARS 2025

作者： Khemnar, Vijaya R. Bendre, Mininath R. Kolekar, Vikas K. Pravara Rural Engineering College Department of Computer Engineering Loni India Vishwakarma Institute of Information Technology Department of Computer Engineering Pune India

ISBN: (纸本)9798331509675

The integration of machine learning (ML) and Internet of Things (IoT) technologies has a scope of improvement in precision farming techniques and revolutionise the agriculture sector. This research paper examines the integration of IoT and ML in soil nutrient monitoring and crop recommendation systems. The aim is to provide a thorough overview of the most recent developments, applications, challenges, and potential future directions. IoT sensors collect and give immediate soil health information, while ML algorithms analyses these data to suggest nutrient deficits and recommend the most suitable crops. This research work evaluates advancements in IoT and Machine learning applications, emphasising their benefits for precision agriculture. This integration aims to improve agricultural production and sustainability via data driven decision making and resource optimisation. At the end research gaps are discussed, along with future research directions. © 2025 ieee.

关键词： Smart agriculture

来源：评论

学校读者我要写书评

暂无评论

The Ethical Implications of AI in Management: Navigating Challenges in Decision-Making

The Ethical Implications of AI in Management: Navigating Cha...

引用

2025 International conference on Pervasive Computational Technologies, ICPCT 2025

作者： Usha, Sruthy Murugan Medatwal, Chandan Purendra Prasad, Vutti Muralidhar, L.B. Rai, Anurag Verma, Amit Saveetha School of Law Simats Chennai77 India Niit University Rajasthan Neemrana India Aims Ibs Business School Department of Management Bangalore India Department of Bms Karnataka Bengaluru India Sharda School of Engineering and Technology Sharda University Department of Computer Science and Applications Greater Noida201310 India Manipal University Jaipur India

ISBN: (纸本)9798331508685

This research uses a tutorial approach to explore the intricate link between artificial intelligence (AI) and management decision-making processes, focussing on the alignment problem, which is a crucial obstacle to ensuring AI systems are in line with moral principles and human values. When defining artificial intelligence, it looks at a variety of methods, including algorithmic methods for machine learning (ML), sentiment analysis, anomaly detection, recommendation systems, expert systems, neural networks, deep learning, data visualisation, and natural language processing. The investigation explores how methodologies for machine learning and predictive analytics may use massive datasets to assist organisations in identifying market dynamics, competitive landscapes and novel patterns. By boosting data-driven decision-making, machine learning (ML) helps businesses proactively adjust to shifting circumstances and take advantage of new opportunities. Additionally, by analysing likely outcomes and their associated probabilities, this study investigates how ML-driven prediction models may lower risks. The investigation also looks at the ethical conundrums that arise when managers utilise ML to make strategic decisions, highlighting the significance of fairness, responsibility, and transparency in decision-making using algorithms. © 2025 ieee.

关键词： Adversarial machine learning

来源：评论

学校读者我要写书评

暂无评论

Machine learning-Enabled Examination Feedback System 3

Machine Learning-Enabled Examination Feedback System

引用

3rd ieee International conference on Industrial Electronics: Developments and Applications, ICIDeA 2025

作者： Ragam, Prashanth Sai, Botcha Taraka Rutesha Mekatotti, Anshita Priyadarshini Ojasvi, Nallamala Jahnavi, Adigoppula Danaveni, Pavithra School of Computer Science & Engineering VIT-AP UNIVERSITY Inavolu AP Vijayawada522237 India Gokaraju Rangaraju Institute of Engineering and Technology Department of Computer Science & Engineering Telangana Hyderabad500090 India

ISBN: (纸本)9798331533205

The project proposes an Examination Feedback System utilizing machine learning to automate the evaluation of subjective answers in educational assessments. The system will analyze student responses and classify them under a number of predefined topics while making sure that the feedback it generates is tailored and specific to the needs of the students. The primary goal is to ensure improved outcomes through enhanced information to students, where improvement is needed, while helping educators identify performance trends. The system is powered by DistilBERT, which is a transformer-based natural language processing model which is known for its efficiency and contextual knowledge. The proposed system differs from previous manual grading systems in that it offers data-driven information while addressing confidentiality issues with a computationally efficient design. Thus, the examination process becomes scalable, accurate, and data-driven in satisfying the demands of upstream education systems with very limited computational infrastructure. © 2025 ieee.

关键词： Natural language processing systems

来源：评论

学校读者我要写书评

暂无评论

Optimization of Power in Power Converters Using Machine learning: An Overview 1

Optimization of Power in Power Converters Using Machine Lear...

引用

1st International conference on Sustainable Energy Technologies and Computational Intelligence, SETCOM 2025

作者： Nayak, Prashant Kumar Swain, Debabrata Panda, Kaibalya Prasad Pandit Deendayal Energy University Department of Computer Science and Engineering Gandhinagar India Pandit Deendayal Energy University School of Energy Technology Department of Electrical Engineering Gandhinagar India

ISBN: (纸本)9798331520540

The rapid evolution in power electronics have brought significant attention to the optimization of power converters, which are essential for efficient energy conversion and management. Traditional techniques for optimizing power converters often struggle with the increasing complexity of modern systems, including nonlinear dynamics, diverse operating conditions, and stringent efficiency requirements. Last few decades, it has been observed that machine learning has evolved as hopeful solution with data driven methodologies to address various challenges. This survey provides the integration of machine learning for optimizing power converter, reviewing new approaches, key procedures and application in different domain. This categorizes different types of machine learning techniques and evaluate their usage in multiple fields like real-time control, fault detection, thermal management and system design. The paper also discusses different datasets, evaluation metrics and implementation framework in field of power converters. Lastly, challenges like computational complexity, data security and real-world deployment are highlighted along with direction towards future research to enhance the integration between machine learning and power converters. The comprehensive review aims to provide insight of researchers carried out to develop an intelligent power system. © 2025 ieee.

关键词： Adversarial machine learning

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：