检索结果-内蒙古大学图书馆

25th international conference on Parallel and Distributed Computing, Applications and Technologies, PDCAT 2024

作者： Zhang, Feng Wang, Jia-Xin Hua, Qiang Dong, Chun-Ru Zhu, Jie Hebei Key Laboratory of Machine Learning and Computational Intelligence College of Mathematics and Information Science Hebei University Baoding China

ISBN: (纸本)9789819642069

To handle model bias from long-tailed image datasets, most existing studies have proposed to strengthen the focus on tail classes. However, they usually overlook the intricacies of sample learning hardness and feature norms’ impact on accuracy, as well as the non-target class contribution. We designed a new loss using logit adjustment. It considers the distribution of the number of class samples, sample learning hardness, and feature norms. Moreover, it combines a complementary entropy regularization term designed for long-tailed data. Extensive experiments on long-tailed datasets confirm the proposed loss’s superiority in enhancing classification accuracy. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.

关键词： Contrastive learning

来源：评论

学校读者我要写书评

暂无评论

Curiosity-driven dual-policy action selection in temporal difference learning for model predictive control

引用

Neural Computing and Applications 2025年 1-17页

作者： Ji, Chang-Hun Choi, Yo-Han Han, Youn-Hee Future Convergence Engineering Korea University of Technology and Education 1600 Chungjeol-ro Chungcheongnam-do Cheonan31253 Korea Republic of

This paper introduces a new reinforcement learning framework that integrates three key methods to optimize action selection into the Temporal Difference Model Predictive Control (TD-MPC) (as reported by Hansen In: international conference on machine learning, 2022), a leading model-based reinforcement learning algorithm, and improve its learning performance in complex environments. First, dual-policy action selection combines the strengths of both model-based and model-free reinforcement learning, enabling more effective and flexible decision-making. Second, policy annealing is introduced to dynamically balance the contributions of these dual policies, preventing excessive exploitation and ensuring robust learning across varying conditions. Third, curiosity-driven exploration is employed to strike an optimal balance between exploration and exploitation, encouraging the agent to explore less familiar states and enhancing overall learning effectiveness. Ultimately, we propose a new scheme, called TD-MPC++, which incorporates the above three key methods into TD-MPC, a leading model-based reinforcement learning algorithm. Through extensive empirical evaluation in the varied environments of the DeepMind Control Suite, we demonstrate that TD-MPC++ achieves not only marked improvements in both sample efficiency and overall training performance when compared to its predecessor, TD-MPC, but also outperforms other state-of-the-art algorithms, including Dreamer V3 and Soft Actor-Critic. Notably, TD-MPC++ achieves superior performance compared to TD-MPC without introducing any additional learning tasks to the models or increasing time complexity. © The Author(s), under exclusive licence to Springer-Verlag London Ltd., part of Springer Nature 2025.

关键词： Reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

4th international conference on machine learning, Image Processing, Network Security and data Sciences, MIND 2022

4th International Conference on Machine Learning, Image Proc...

引用

4th international conference on machine learning, Image Processing, Network Security and data Sciences, MIND 2022

ISBN: (纸本)9783031243660

The proceedings contain 64 papers. The special focus in this conference is on machine learning, Image Processing, Network Security and data Sciences. The topics include: Sensorless Control Algorithm of Permanent Magnet Synchronous Motor on Account of Neural Network;a Techno Aid to Ease in e-Rehabilitation;a Novel Approach to Analyse Lung Cancer Progression and Metastasis Using Page Rank Technique;homomorphic Encryption of Neural Networks;an Empirical Study to Enhance the Accuracy of an Ensemble learning Model for Crop Recommendation System by Using Bit-Fusion Algorithm;bayesian learning Model for Predicting Stability of System with Nonlinear Characteristics;Novel ABC: Aspect Based Classification of Sentiments Using Text Mining for COVID-19 Comments;topic Modeling, Sentiment Analysis and Text Summarization for Analyzing News Headlines and Articles;bioBodyComp: A machine learning Approach for Estimation of Percentage Body Fat;a Computational Approach to Identify Normal and Abnormal Persons Gait Using Various machine learning and Deep learning Classifier;wearable Technology for Early Detection of Hyperthermia Using machine learning;intelligent Evaluation Framework of English Composition Based on Intelligent Algorithm;multimedia English Teaching System Based on Computer Information Technology;Correlation Analysis of Central Bank Communication Behavior and Monetary Policy Independence Based on VR Technology and machine learning;power Demand data Analysis and Recovery for Management of Power Distribution Systems;Optimization Design of Green Building Landscape Space Environment Based on LM-BP Algorithm;design of Computational Thinking Intelligent Training System Under Big data Technology;creative Graphic Design System Based on Multi-objective Firefly Algorithm;next Generation Ultra-sensitive Surface Plasmon Resonance Biosensors.

关键词：

来源：评论

学校读者我要写书评

暂无评论

4th international conference on machine learning, Image Processing, Network Security and data Sciences, MIND 2022

4th International Conference on Machine Learning, Image Proc...

引用

4th international conference on machine learning, Image Processing, Network Security and data Sciences, MIND 2022

ISBN: (纸本)9783031243516

关键词：

来源：评论

学校读者我要写书评

暂无评论

Comparative Analysis of Traditional and Modern Techniques to Study the Influence of Health on Travel Mode Choice to Work Using 2022 NHTS 20th

Comparative Analysis of Traditional and Modern Techniques to...

引用

20th Scientific and Technical conference on Transport Systems Theory and Practice, TSTP 2024

作者： Ali, Mujahid Macioszek, Elżbieta Faculty of Transport and Aviation Engineering Department of Transport System Traffic Engineering and Logistics Silesian University of Technology Katowice Poland

ISBN: (纸本)9783031857508

Several studies have been conducted to investigate the influence of transport-related daily activities on health parameters using conventional methods. These traditional methods are based on several assumptions that produce bias in results. Besides, travel mode choice (TMC) data is complex and imbalanced whereas traditional methods fail to handle imbalances and complex datasets. Therefore, the current study aims to use a reciprocal approach to study the influence of health parameters on work TMC using both traditional and modern techniques. Some of the variables - based on the scope of the study, are used from the 2022 National Household Travel Survey (NHTS), California, to study the correlation between health parameters and work TMC. RStudio is used for multivariate regression analysis, whereas several machine learning (ML) algorithms were developed to compare their results based on the classification matrix and to suggest the best prediction model for work TMC. The statistical analysis shows that all the models are statistically significant with a p-value of less than 0.005 and the models are extremely acceptable range with an R2 value of over 30%. Those individuals with disabilities - using manual wheelchairs were the most significant feature that influenced work TMC which shows that health influences work TMC. The current professes those modern techniques outperformed traditional methods where SVM followed by RF shows the best accuracy and precision. SVM shows the best predictions with an accuracy of 47%, whereas RF with an accuracy of 44% and a precision of 36%. The current study helps urban planners and policymakers to provide a more convenient transportation system for disabled people and enhance the reliability, frequency, and safety of public transportation. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.

关键词： disabilities health parameters machine learning algorithms sustainable transportation systems Travel mode choice

来源：评论

学校读者我要写书评

暂无评论

RF Challenge: The data-Driven Radio Frequency Signal Separation Challenge

IEEE Open Journal of the Communications Society

引用

IEEE Open Journal of the Communications Society 2025年 6卷 4083-4100页

作者： Lancho, Alejandro Weiss, Amir Lee, Gary C. F. Jayashankar, Tejas Kurien, Binoy G. Polyanskiy, Yury Wornell, Gregory W. Universidad Carlos III de Madrid Signal Theory and Communications Leganés28911 Spain Gregorio Marañón Health Research Institute Madrid28007 Spain Bar-Ilan University Ramat Gan5290002 Israel Institute for Infocomm Research Communications and Networks Fusionopolis 138632 Singapore Massachusetts Institute of Technology Electrical Engineering and Computer Science Department CambridgeMA02139 United States MIT Lincoln Laboratory LexingtonMA02421 United States

We address the critical problem of interference rejection in radio-frequency (RF) signals using a data-driven approach that leverages deep-learning methods. A primary contribution of this paper is the introduction of the RF Challenge, which is a publicly available, diverse RF signal dataset for data-driven analyses of RF signal problems. Specifically, we adopt a simplified signal model for developing and analyzing interference rejection algorithms. For this signal model, we introduce a set of carefully chosen deep learning architectures, incorporating key domain-informed modifications alongside traditional benchmark solutions to establish baseline performance metrics for this intricate, ubiquitous problem. Through extensive simulations involving eight different signal mixture types, we demonstrate the superior performance (in some cases, by two orders of magnitude) of architectures such as UNet and WaveNet over traditional methods like matched filtering and linear minimum mean square error estimation. Our findings suggest that the data-driven approach can yield scalable solutions, in the sense that the same architectures may be similarly trained and deployed for different types of signals. Moreover, these findings further corroborate the promising potential of deep learning algorithms for enhancing communication systems, particularly via interference mitigation. This work also includes results from an open competition based on the RF Challenge, hosted at the 2024 IEEE international conference on Acoustics, Speech, and Signal Processing (ICASSP'24). © 2020 IEEE.

关键词： Matched filters

来源：评论

学校读者我要写书评

暂无评论

Characterization of Water Consumers in Urban Areas Based on data Visualization Techniques 19th

Characterization of Water Consumers in Urban Areas Based on ...

引用

19th international conference on Hybrid Artificial Intelligence Systems

作者： Rubinos, Manuel Arcano-Bea, Paula Diaz-Longueira, Antonio Timiraos, Miriam Michelena, Alvaro Zayas-Gato, Francisco Univ A Coruna Dept Ind Engn CTC CITIC Avda 19 Febrero S-N Ferrol 15405 A Coruna Spain Fdn Inst Tecnol Galicia Natl Technol Ctr Dept Water Technol Galicia 15003 A Coruna Spain

ISBN: (纸本)9783031741852;9783031741869

The current growth in population density has increased the demand for potable water and aggravated water scarcity in reservoirs, especially in urban and metropolitan areas. This problem, coupled with other aspects such as climate change and rain irregularity, has proven that current water management techniques, which rely on human operator knowledge, are insufficient and inefficient. Observing the development of data Science and machine learning techniques, these intelligent algorithms seem to be an interesting approach that could support water management operations by predicting water usage depending on several factors. A first step involves characterizing user consumption behaviors by applying feature extraction and data visualization techniques to discover similarities and common patterns among users.

关键词： Urban water management Water consumption User characterization Feature extraction data visualization t-SNE UMAP

来源：评论

学校读者我要写书评

暂无评论

Prediction of Benign and Malignant Thyroid Nodules Based on machine learning 24

Prediction of Benign and Malignant Thyroid Nodules Based on ...

引用

2024 international conference on Virtual Reality, Image and Signal Processing, ICVISP 2024

作者： Zhang, Jiaoxia Chen, Sihui Beijing Institute of Technology Zhuhai China

ISBN: (纸本)9798400710926

Thyroid nodules are a common endocrine organ disease, but up to 30% of thyroid nodules cannot be accurately classified as benign or malignant through cell pathology. In order to reduce the number of unnecessary surgeries for patients with benign nodules, this article predicts the benign and malignant nature of thyroid nodules. Firstly, fuzzy matching was performed on 2650 sample data from 1143 patients, and missing values were processed for 87 protein matrix indicators related to thyroid nodules. Secondly, a feature extraction technique based on genetic algorithm was adopted to select the indicators that are more likely to display malignant thyroid nodules from the preliminary extracted indicators, and random forest model, SVM model, XGBoost model, logistic model, and logistic XGBoost model were established. Finally, through comparison, the logistic based XGBoost model has a recognition accuracy of 80.2% and an AUC value of 87.2%, indicating the best classification performance. © 2024 Copyright held by the owner/author(s).

关键词： Endocrinology

来源：评论

学校读者我要写书评

暂无评论

23rd international conference on Intelligent data engineering and Automated learning, IDEAL 2022

23rd International Conference on Intelligent Data Engineerin...

引用

23rd international conference on Intelligent data engineering and Automated learning, IDEAL 2022

ISBN: (纸本)9783031217524

The proceedings contain 52 papers. The special focus in this conference is on Intelligent data engineering and Automated learning. The topics include: Benchmarking data Augmentation Techniques for Tabular data;deep learning Based Predictive Analytics for Decentralized Content Caching in Hierarchical Edge Networks;explanations of Performance Differences in Segment Lining for Tunnel Boring machines;On Autonomous Drone Navigation Using Deep learning and an Intelligent Rainbow DQN Agent;an Intelligent Decision Support System for Road Freight Transport;endowing Intelligent Vehicles with the Ability to Learn User’s Habits and Preferences with machine learning Methods;duplication Scheduling with Bottom-Up Top-Down Recursive Neural Network;towards a Low-Cost Companion Robot for Helping Elderly Well-Being;zero-Shot Knowledge Graph Completion for Recommendation System;synonym-Based Essay Generation and Augmentation for Robust Automatic Essay Scoring;the Covid-19 Influence on the Desire to Stay at Home: A Big data Architecture;distance-Based Delays in Echo State Networks;eduBot: A Proof-of-Concept for a High School Motivational Agent;A Simulation Model for Predicting the Spread of COVID-19 Virus;ICU Mortality Prediction Using Long Short-Term Memory Networks;intelligent learning Rate Distribution to Reduce Catastrophic Forgetting in Transformers;how Image Retrieval and Matching Can Improve Object Localisation on Offshore Platforms;Ethereum Investment Based on LSTM and GRU Forecast;Generating a European Portuguese BERT Based Model Using Content from *** Archive;a Vision Transformer Enhanced with Patch Encoding for Malware Classification;characterizing Cardiovascular Risk Through Unsupervised and Interpretable Techniques;association Rules Mining for Reducing Items from Emotion Regulation Questionnaires;explainable Artificial Intelligence for Improved Modeling of Processes;efficient Sensor Selection for Individualized Prediction Based on Biosignals;understanding the Classes

关键词：

来源：评论

学校读者我要写书评

暂无评论

An LSTM-based detection model for breathing cracks with multiple damage positions and degrees of beam-like bridges under vehicle loads

引用

STRUCTURAL HEALTH MONITORING-AN international JOURNAL 2025年

作者： Wang, Cheng Yang, Zhen Gao, Kang Xia, Ye Li, Yi Ng, Ching-Tai Wu, Gang Southeast Univ Key Lab Concrete & Prestressed Concrete Struct Minist Educ 2 Dongnandaxue Rd Nanjing 211189 Peoples R China Natl & Local Joint Engn Res Ctr Intelligent Constr Nanjing Peoples R China Tongji Univ Dept Bridge Engn Shanghai Peoples R China State Key Lab Disaster Reduct Civil Engn Shanghai Peoples R China Univ Adelaide Sch Architecture & Civil Engn Adelaide SA Australia

The cracks of bridges are not always completely open during vibrations but undergo an open and closed cycle mode under moving vehicles. Therefore, this study proposes a long short-term memory (LSTM)-based detection model for breathing cracks with multiple damage positions and degrees of beam-like bridges for the first time. The contact point displacement variation (CPDV), serving as an evaluation indicator of the bridge damage, is calculated by analyzing the acceleration history data of the vehicle that has passed over a randomly damaged bridge. At the same time, CPDV is also used as the input variable of an LSTM neural network to monitor the damage location and degree of the bridge. By employing the finite element simulation of the bridge half-vehicle model, a dataset was created for the training and prediction of LSTM. The numerical results show that the proposed method can accurately identify complex breathing crack information at different vehicle speeds, and it is also robust to road roughness. In addition, a laboratory experiment was conducted to verify the proposed method. The experimental results show that CPDV is sensitive to bridge cracks and LSTM can realize higher damage prediction.

关键词： Breathing crack detection contact point displacement variation machine learning long short-term memory drive-by bridge damage inspection

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：