检索结果-内蒙古大学图书馆

11th international symposium on computational intelligence and Industrial Applications, ISCIIA 2024

作者： Mersha, Bemnet Wondimagegnehu Dai, Yaping Hirota, Kaoru School of Automation Beijing Institute of Technology Beijing100081 China Beijing Institute of Technology Zhuhai Tangjiawan Guangdong Zhuhai519088 China

ISBN: (纸本)9789819647521

To ensure reliable predictions for various applications, it is essential to understand the impact of Gaussian and non-Gaussian noise on data-driven modeling. Gaussian noise has been a convenient assumption for model development, but real-world scenarios sometimes defy Gaussian assumptions. Our research compared the performances of the direct and parametric data-driven modeling methods with Gaussian and non-Gaussian sensor noise. The direct data-driven modeling methods that are evaluated are direct data-driven methods with pulse input-output dataset and direct data-driven methods with step input-output dataset. The parametric models that are evaluated are the Output Error (OE), Autoregressive with Moving Average and Exogenous Input (ARMAX), Autoregressive with Exogenous Input (ARX) and Box-Jenkins model (BJ). The result shows that the performance of the direct and the parametric data-driven modeling methods deteriorates under non-Gaussian noise. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.

关键词： Gaussian noise (electronic)

来源：评论

学校读者我要写书评

暂无评论

Artificial intelligence in assisted reproductive technology: separating the dream from reality

引用

Reproductive BioMedicine Online 2025年第4期50卷 104855-104855页

作者： Cohen, Jacques Silvestri, Giuseppe Paredes, Omar Martin-Alcala, Hector E. Chavez-Badiola, Alejandro Alikani, Mina Palmer, Giles A. Conceivable Life Sciences New York NY United States International IVF Initiative New York NY United States IVF 2.0 Ltd London United Kingdom Althea Science New York NY United States Biodigital Innovation Laboratory Department of Translational Bioengineering Centro Universitario de Ciencias Exactas e Ingenierías Universidad of Guadalajara Mexico Mexico New Hope Clinic Guadalajara Mexico Alpha Scientists in Reproductive Medicine London United Kingdom Institute of Life IASO Hospital Athens Greece

This paper critically reviews the role of artificial intelligence (AI) in assisted reproductive technology (ART), a nascent field that has emerged over the last decade. While AI holds immense promise for enhancing IVF efficiency, standardization, and outcomes, its current trajectory reveals significant challenges. Much of the recent literature presents variations on established methodologies rather than groundbreaking advancements, with many studies lacking clear clinical applications or outcome-driven validations. Moreover, the growing enthusiasm for AI in ART is often accompanied by undue hype that obscures its realistic potential and fosters inflated expectations. Despite these limitations, AI-driven innovations such as advanced image analysis, personalized protocols, and automation of embryology workflows are beginning to show value. Machine learning algorithms and robotics may help address inefficiencies, alleviate staff shortages, and improve decision-making in the IVF laboratory. However, progress is tempered by drawbacks including ethical concerns, limited transparency in AI systems, and regulatory impediments. Data-sharing barriers in our field hinder AI tool development significantly. Energy-intensive computational processes and expanding data centers also raise sustainability concerns, underscoring the need for environmentally responsible development. As the field evolves, it must emphasize rigorous validation, collaborative data frameworks, and alignment with the needs of ART practitioners and patients. © 2025 The Author(s)

关键词： Artificial intelligence Assisted reproduction automation Embryology Machine learning

来源：评论

学校读者我要写书评

暂无评论

Analysis of Machine Learning Models for Stroke Prediction with Emphasis on Hyperparameter Tuning Techniques 11th

Analysis of Machine Learning Models for Stroke Prediction wi...

引用

11th international symposium on computational intelligence and Industrial Applications, ISCIIA 2024

作者： Hasan, Sakib Islam, Alamgir Islam, Tanjin Ma, Hongbin School of Information and Electronics Beijing Institute of Technology Beijing100811 China School of Automation Beijing Institute of Technology Beijing100811 China School of Automation Jiangsu University of Science and Technology Zhenjiang China

ISBN: (纸本)9789819647552

Stroke remains a significant global cause of death and disability, necessitating early and accurate prediction models for prompt intervention. This study contrasts the performance of Support Vector Machine (SVM) and Random Forest (RF) models to enhance stroke prediction approaches. Emphasizing the critical role of hyper parameter adjustment in improving model efficiency, two tuning methods—Grid Search Cross-Validation (GS-CV) and Randomized Search Cross-Validation (RS-CV)—are investigated. Data prepossessing utilizes a data set from the Medical Clinic of Bangladesh, comprising 5,110 patient records. Imbalanced data is addressed through the Synthetic Minority Over-sampling Technique (SMOTE). Despite being good at predicting accuracy, SVM with RS-CV tuning is more accurate, achieving a 96% accuracy than RF with GS-CV tuning that achieves 92% accuracy. Such outcomes highlight the significance of choosing proper hyperparameter tuning techniques and ML models for stroke prediction. They also imply an outlet for use in healthcare contexts concerning early identification and prophylactic steps. This comparison study adds to the current debate about machine learning in medical prediction, focusing on the methodological aspects critical to constructing reliable and effective predictive systems. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.

关键词： Patient treatment

来源：评论

学校读者我要写书评

暂无评论

Two Time Scale Partial Unknown Dynamics System Tracking Control Based on Off-Policy Inverse Reinforcement Learning 11th

Two Time Scale Partial Unknown Dynamics System Tracking Cont...

引用

11th international symposium on computational intelligence and Industrial Applications, ISCIIA 2024

作者： Shen, MinYin Liu, Fei Institute of Automation Jiangnan University Wuxi214122 China

ISBN: (纸本)9789819647552

This article, integrating the singular perturbation technique with inverse reinforcement learning, proposes a novel linear two time scale system tracking control method grounded in off-policy inverse reinforcement learning. This method addresses the challenge of unknown cost functions prevalent in industrial processes. First, the singular perturbation method is leveraged to decompose the original problem into fast and slow subsystem issues. Without manually designing a cost function, the method learns from known optimal behavioral data by reconstructing cost functions tailored to each subsystem, enabling the system to mimic optimal behaviors. Then, for the fast time scale system, a model-based inverse reinforcement learning method is adopted, while for the slow time scale system, a model-free off-policy inverse reinforcement learning strategy is employed, which reconstructs the system’s cost function solely using measured expert behavioral data inputs. Finally, using a mixed separation thickening industrial process to illustrate the effectiveness of this method in two time scale tracking. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.

关键词： Cost functions

来源：评论

学校读者我要写书评

暂无评论

SwinLip: An Efficient Visual Speech Encoder for Lip Reading Using Swin Transformer

引用

Neurocomputing 2025年 639卷

作者： Park, Young-Hu Park, Rae-Hong Park, Hyung-Min Department of Artificial Intelligence Sogang University Seoul04107 Korea Republic of Department of Electronic Engineering Sogang University Seoul04107 Korea Republic of

This paper presents an efficient visual speech encoder for lip reading. While most recent lip reading studies have been based on the ResNet architecture and have achieved significant success, they are not sufficiently suitable for efficiently capturing lip reading features due to high computational complexity in modeling spatio-temporal information. Additionally, using a complex visual model not only increases the complexity of lip reading models but also induces delays in the overall network for multi-modal studies (e.g., audio–visual speech recognition, speech enhancement, and speech separation). To overcome the limitations of Convolutional Neural Network (CNN)-based models, we apply the hierarchical structure and window self-attention of the Swin Transformer to lip reading. We configure a new lightweight scale of the Swin Transformer suitable for processing lip reading data and present the SwinLip visual speech encoder, which efficiently reduces computational load by integrating modified Convolution-augmented Transformer (Conformer) temporal embeddings with conventional spatial embeddings in the hierarchical structure. Through extensive experiments, we have validated that our SwinLip successfully improves the performance and inference speed of the lip reading network when applied to various backbones for word and sentence recognition, reducing computational load. In particular, our SwinLip demonstrated robust performance in both English LRW and Mandarin LRW-1000 datasets and achieved state-of-the-art performance on the Mandarin LRW-1000 dataset with less computation compared to the existing state-of-the-art model. © 2025 Elsevier B.V.

关键词： Speech enhancement

来源：评论

学校读者我要写书评

暂无评论

Multi-agent Reinforcement Learning for Sparse Reward Tasks Using Incremental Goal Enhanced Method 11th

Multi-agent Reinforcement Learning for Sparse Reward Tasks ...

引用

11th international symposium on computational intelligence and Industrial Applications, ISCIIA 2024

作者： Han, Minglei Guo, Zhentao Sun, Licheng Ding, Ao Wang, Tianhao Zhao, Guiyu Ma, Hongbin School of Automation Beijing Institute of Technology Beijing100081 China National Key Lab of Autonomous Intelligent Unmanned Systems Beijing100081 China

ISBN: (纸本)9789819647552

As the application of artificial intelligence continues to expand, complex decision-making problems such as multi-player gaming, multi-robot planning and multi-vehicle controlling have become new challenges for machine intelligence. MARL which concentrates on learning the optimal strategies of multiple agents that coexist in a shared environment, is a valid method to solve multi-agent decision-making challenges. Among MARL Algorithms, the MAPPO algorithm has won the favor of machine learning community due to its superb performance. However, the original MAPPO algorithm suffers from sparse reward issues. To overcome the sparse rewards problem and achieve sufficient learning in complex task, this paper proposes a IGE-MAPPO which uses a IGM that generates a variable-density and bi-domain reward signal, and conducts experiments on SMAC. The results show that the IGE-MAPPO algorithm can adapt to a variety of complex environment and has improved performance compared with other typical MARL algorithms. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.

关键词： Reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

Efficient and Accurate Point Cloud Registration with Sparsepoint Transformer for Landslide Detection 11th

Efficient and Accurate Point Cloud Registration with Spars...

引用

11th international symposium on computational intelligence and Industrial Applications, ISCIIA 2024

作者： Zhao, Guiyu Wang, Xin Guo, Zhentao Ma, Hongbin School of Automation Beijing Institute of Technology Beijing China National Key Lab of Autonomous Intelligent Unmanned Systems Beijing Institute of Technology Beijing China

ISBN: (纸本)9789819647552

We present an efficient and accurate point cloud registration method that has been successfully applied to real-world landslide detection tasks. Existing point cloud registration methods often suffer from either low accuracy or high computational complexity, rendering them impractical for real-world applications. In this paper, we propose an efficient and accurate SparsePoint Transformer framework for point cloud registration, which directly applies an attention mechanism to sparse points, significantly enhancing computational efficiency. Additionally, we introduce a cascaded feature aggregation encoder to enrich the contextual details in point clouds, thereby improving registration accuracy. To further adapt our framework for landslide detection, we propose the point cloud registration for landslide detection (PCR4LD) framework, built on the SparsePoint Transformer pipeline. This framework first addresses vegetation interference with two proposed solutions. Subsequently, it employs ICP-based pose refinement for further pose refinement, ensuring accurate landslide detection. Finally, region merging and filtering are applied to identify the landslide-affected regions. Our method not only achieves state-of-the-art registration results on the public KITTI dataset with a significant speedup but also demonstrates outstanding performance in real-world landslide detection tasks, significantly outperforming other methods. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.

关键词： Image registration

来源：评论

学校读者我要写书评

暂无评论

Multi-objective Evolutionary Algorithm Based Graph Neural Network Architecture Search 11th

Multi-objective Evolutionary Algorithm Based Graph Neural Ne...

引用

11th international symposium on computational intelligence and Industrial Applications, ISCIIA 2024

作者： He, Lianyi Liu, Xiaobo Xiang, Hongbo Wang, Guangjun School of Automation China University of Geosciences Wuhan430074 China Hubei Key Laboratory of Advanced Control and Intelligent Automation for Complex Systems Wuhan430074 China Engineering Research Center of Intelligent Technology for Geo-Exploration Ministry of Education Wuhan430074 China Key Laboratory of Geological Survey and Evaluation of Ministry of Education China University of Geosciences Wuhan430074 China

ISBN: (纸本)9789819647521

Graph Neural Networks (GNN) has become a powerful graph data processing method, which has been widely used in node classification, link prediction, and other graph analysis tasks. Due to the diversity and complexity of graph structures and information propagation, as well as the handling of heterogeneous graphs, etc., the design of GNNs presents many challenges. The existing methods have high classification accuracy, but their structures are very complex. In this paper, a graph neural network architecture search framework named MO-GNN, which is based on multi-objective evolutionary algorithm is proposed. Furthermore, a weight sharing strategy MO-GNN-WS is proposed to reduce the resource consumption caused by weight training of different architectures. To verify the performance of the proposed algorithm, experiments on four popular graph datasets are used for transduction and induction tasks. The experimental results show that the MO-GNN-WS algorithm outperforms the most advanced neural network architecture search methods in terms of classification accuracy and resource consumption. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.

关键词： Data Sharing

来源：评论

学校读者我要写书评

暂无评论

Research of Sequential Recommendation Algorithm Based on Contrastive Learning and Causal Learning 11th

Research of Sequential Recommendation Algorithm Based on Con...

引用

11th international symposium on computational intelligence and Industrial Applications, ISCIIA 2024

作者： Wang, Tong Dai, Yaping Shao, Shuai School of Automation Beijing Institute of Technology Beijing100081 China

ISBN: (纸本)9789819647521

In sequential recommender systems, two primary challenges are the long-tailed distribution of data and data distribution bias. To effectively address these issues, a Contrastive and Causal Learning Algorithm for Sequential Recommendation (C2ASeRec) has been proposed. The algorithm enhances the training efficacy of sequential recommendation models and boosts their performance by introducing environment partition and reweighting, regularization term constraint based on causal learning, and methods to enhance uniformity of representation. These innovations mitigate the performance degradation previously caused by data distribution bias. By concurrently incorporating causal learning-based regularization constraints and representation uniformity enhancement techniques, C2ASeRec demonstrates both universality and robustness across different environment partitioning principles, enabling superior performance in complex real-world scenarios. Experimental results indicate that C2ASeRec achieves outstanding outcomes in addressing data distribution bias. In terms of key performance metrics such as hit rate and normalized discounted cumulative gain, our algorithm significantly surpasses seven previous methods, showcasing exceptional advanced performance. © 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

Graph Attention Convolutional Neural Network Combined with Transformer Classification Method for Small Sample Datasets 11th

Graph Attention Convolutional Neural Network Combined with T...

引用

11th international symposium on computational intelligence and Industrial Applications, ISCIIA 2024

作者： Lei, Rongzhen Yang, Fan Ren, Yanhui School of Instrument Science and Opto-Electronics Engineering Beijing Information Science and Technology University Beijing100192 China Department of Automation Tsinghua University Beijing10084 China

ISBN: (纸本)9789819647521

In view of the shortcomings of deep learning methods in modeling small sample fault data, this paper proposes a fault classification method based on Graph Attention Convolutional Neural Network (GATCN) combined with Transformer, which can be well adapted to small sample fault data sets. Firstly, the multi-dimensional sensor data is converted into a multi-component graph representation, and the weights of the edges are updated by the attention mechanism between the nodes of the graph to represent the topological structure between different nodes, to extract the correlation between the multi-component parameters. Then, the Transformer module learn the extracted spatial features to capture the temporal feature relationships of the time series data. Finally, the classification layer based on Softmax is used to classify faults. The model is demonstrated on the TEP data set and the fault datasets from three satellite subsystems. The results show that our model is superior for small sample fault data sets. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.

关键词： Convolutional neural networks

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：