Applied Intelligence最新文献_第4页

Enhancing model learning in reinforcement learning through Q-function-guided trajectory alignment 通过q函数引导轨迹对齐增强强化学习中的模型学习

IF 3.4 2区计算机科学

Applied Intelligence Pub Date : 2025-04-29 DOI: 10.1007/s10489-024-06083-9

Xin Du, Shan Zhong, Shengrong Gong, Yali Si, Zhenyu Qi

{"title":"Enhancing model learning in reinforcement learning through Q-function-guided trajectory alignment","authors":"Xin Du, Shan Zhong, Shengrong Gong, Yali Si, Zhenyu Qi","doi":"10.1007/s10489-024-06083-9","DOIUrl":"10.1007/s10489-024-06083-9","url":null,"abstract":"<div><p>Model-based reinforcement learning (MBRL) methods hold great promise for achieving excellent sample efficiency by fitting a dynamics model to previously observed data and leveraging it for RL or planning. However, the resulting trajectories may diverge from actual-world trajectories due to the accumulation of errors in multi-step model sampling, particularly for longer horizons. This undermines the performance of MBRL and significantly affects sample efficiency. Therefore, we present a trajectory alignment capable of aligning simulated trajectories with their real counterparts from any initial random state and with adaptive length, enabling the preparation of paired real-simulated samples to minimize compounding errors. Additionally, we design a Q-function function to estimate Q values for the paired real-simulated samples. The simulated samples whose Q-value difference from the real ones surpasses a given threshold will be discarded, thus preventing the model from over-fitting to erroneous samples. Experimental results demonstrate that both trajectory alignment and Q-function guided sample filtration contribute to improving policy and sample efficiency. Our method surpasses previous state-of-the-art model-based approaches in both sample efficiency and asymptotic performance across a series of challenging control tasks. The code is open source and available at https://github.com/duxin0618/qgtambpo.git.</p></div>","PeriodicalId":8041,"journal":{"name":"Applied Intelligence","volume":"55 10","pages":""},"PeriodicalIF":3.4,"publicationDate":"2025-04-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143883736","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

LiteSpiralGCN: Lightweight 3D hand mesh reconstruction via spiral graph convolution LiteSpiralGCN：基于螺旋图卷积的轻量级3D手部网格重建

IF 3.4 2区计算机科学

Applied Intelligence Pub Date : 2025-04-29 DOI: 10.1007/s10489-025-06585-0

Yiteng Wang, Minqi Li, Kaibing Zhang, Xiangjian He

引用次数: 0

Adaptive robust cost-sensitive online classification algorithm for class-imbalanced datasets 类不平衡数据集的自适应鲁棒代价敏感在线分类算法

IF 3.4 2区计算机科学

Applied Intelligence Pub Date : 2025-04-29 DOI: 10.1007/s10489-025-06567-2

Xian Shan, Jinyu You, Xiaoying Li, Zheshuo Zhang, Yu Xie

{"title":"Adaptive robust cost-sensitive online classification algorithm for class-imbalanced datasets","authors":"Xian Shan, Jinyu You, Xiaoying Li, Zheshuo Zhang, Yu Xie","doi":"10.1007/s10489-025-06567-2","DOIUrl":"10.1007/s10489-025-06567-2","url":null,"abstract":"<div><p>With the continuous development of machine learning technology, classification has become increasingly important in various fields, such as disease detection, user analysis, etc. However, traditional classification algorithms frequently encounter challenges such as class imbalances, noise and outliers, and large-scale dynamic data processing, which limit their performance in practical applications. This study presents an enhanced adaptive robust cost-sensitive online classification algorithm that dynamically adjusts the penalty coefficient according to the distribution characteristics of the data stream and the algorithm’s performance, in combination with an online learning strategy, to improve the model’s robustness in dealing with dynamic data streams, class imbalance, and noise or outliers. A series of numerical experiments and real-world applications have validated that the new algorithm can significantly enhance classification accuracy while maintaining computational efficiency. Notably, the algorithm demonstrates promising application potential in practical problems such as credit card default detection.</p></div>","PeriodicalId":8041,"journal":{"name":"Applied Intelligence","volume":"55 10","pages":""},"PeriodicalIF":3.4,"publicationDate":"2025-04-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143883573","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

NS-FUO: Fourier U-type operator based on nested structure NS-FUO：基于嵌套结构的傅里叶u型算子

IF 3.4 2区计算机科学

Applied Intelligence Pub Date : 2025-04-29 DOI: 10.1007/s10489-025-06552-9

Jingjian Chen, Jie Nie, Ning Song, Min Ye, Zhiqiang Wei

{"title":"NS-FUO: Fourier U-type operator based on nested structure","authors":"Jingjian Chen, Jie Nie, Ning Song, Min Ye, Zhiqiang Wei","doi":"10.1007/s10489-025-06552-9","DOIUrl":"10.1007/s10489-025-06552-9","url":null,"abstract":"<div><p>For partial differential equations (PDE), neural operators can learn the mapping of input and output functions in infinite dimensional spaces by introducing kernel functions into linear transformations. Fourier neural operator (FNO) is a very representative neural operator, which filters out the high-frequency noise in PDE mainly through low frequency dominated Fourier space truncation, and can solve PDE with high precision and high efficiency. However, for some complex high-dimensional PDE, FNO and other algorithms usually have the problem of incomplete filtering out high-frequency noise, which will affect the solution accuracy. To filter out high-frequency noise more thoroughly and further improve the precision, we propose NS-FUO: Fourier U-type Operator Based on Nested Structure. Firstly, NS-FUO adds MLP to each Fourier layer to extract the nonlinear features of PDE in depth. Then, NS-FUO adds UNet to each Fourier layer to extract the multi-layer condition features of PDE in depth. Finally, NS-FUO adds nested UNet after the last Fourier layer to fuses the original input features of PDE with the filtered output features. The experimental results show that compared with 15 PDE intelligent methods such as FNO, U-FNO, LSM, etc, NS-FUO has the highest accuracy for solving three solid PDEs and four fluid PDEs, and achieves an average accuracy improvement of 11.9% compared with the previous best method LSM.</p></div>","PeriodicalId":8041,"journal":{"name":"Applied Intelligence","volume":"55 10","pages":""},"PeriodicalIF":3.4,"publicationDate":"2025-04-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143883735","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

LiDAR-based perception system for logistics in industrial environments 基于激光雷达的工业环境物流感知系统

IF 3.4 2区计算机科学

Applied Intelligence Pub Date : 2025-04-28 DOI: 10.1007/s10489-025-06528-9

Martín Palos, Irene Cortés, Ángel Madridano, Francisco Navas, Carmen Barbero, Vicente Milanés, Fernando García

{"title":"LiDAR-based perception system for logistics in industrial environments","authors":"Martín Palos, Irene Cortés, Ángel Madridano, Francisco Navas, Carmen Barbero, Vicente Milanés, Fernando García","doi":"10.1007/s10489-025-06528-9","DOIUrl":"10.1007/s10489-025-06528-9","url":null,"abstract":"<div><p>Autonomous vehicles in logistics and industrial environments demand robust and efficient perception systems. This study presents a LiDAR-based perception system designed for such environments, focusing on real-time deterministic obstacle detection and tracking with limited computational power. The proposed multi-stage approach leverages 3D data from LiDAR sensors. First, ground removal is performed to filter out static ground points. Then, a filtering step is applied using precomputed maps of the navigation area to filter out static zones from the LiDAR point clouds. After, object segmentation distinguishes structural elements from potential obstacles, followed by clustering and Principal Component Analysis (PCA) to accurately estimate obstacle pose and volume. An obstacle-tracking method ensures continuous monitoring over time. Extensive experiments in realistic logistics and industrial scenarios have been performed, comparing the proposed approach to state-of-the-art deep-learning-based methods, demonstrating the system’s high performance in both accuracy and efficiency.</p></div>","PeriodicalId":8041,"journal":{"name":"Applied Intelligence","volume":"55 10","pages":""},"PeriodicalIF":3.4,"publicationDate":"2025-04-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1007/s10489-025-06528-9.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143879585","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

MMCA: Multi-stage multi-order context aggregation framework for LDCT denoising 用于LDCT去噪的多阶段多阶上下文聚合框架

IF 3.4 2区计算机科学

Applied Intelligence Pub Date : 2025-04-28 DOI: 10.1007/s10489-025-06553-8

Jianfang Li, Li Wang, Shengxiang Wang, Yakang Li, Fazhi Qi

{"title":"MMCA: Multi-stage multi-order context aggregation framework for LDCT denoising","authors":"Jianfang Li, Li Wang, Shengxiang Wang, Yakang Li, Fazhi Qi","doi":"10.1007/s10489-025-06553-8","DOIUrl":"10.1007/s10489-025-06553-8","url":null,"abstract":"<div><p>Low-dose computed tomography (LDCT) is widely used to reduce patient radiation exposure, but this reduction often comes at the cost of increased noise in the CT images. Although various deep learning-based methods have been developed for LDCT denoising, most struggle to balance local perception and global contextual capture, thus failing to highlight valuable expressions. This paper presents a multi-stage multi-order context aggregation learning framework designed for high-resolution feature map. The framework combines local perception with adaptive context aggregation to improve performance. Each stage employs the macro-architecture of a vision transformer and integrates edge-enhancement features. Initially, the input passes through feature embedding blocks, followed by the stacking of multiple multi-order context aggregation modules to enable efficient feature interaction. The context aggregation modules effectively generate more discriminative representations from features that incorporate edge information. Extensive experiments on two publicly available LDCT denoising datasets demonstrate that our method surpasses state-of-the-art models. Notably, our method strikes a better balance between network efficiency and denoising performance. The code will be made publicly available on https://code.ihep.ac.cn/lijf/MMCA.</p></div>","PeriodicalId":8041,"journal":{"name":"Applied Intelligence","volume":"55 10","pages":""},"PeriodicalIF":3.4,"publicationDate":"2025-04-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143879638","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

TFP-mixer: A lightweight time and frequency combining model for multivariate long-term time series forecasting TFP-mixer：用于多变量长期时间序列预测的轻量级时间和频率组合模型

IF 3.4 2区计算机科学

Applied Intelligence Pub Date : 2025-04-28 DOI: 10.1007/s10489-025-06562-7

Zhaodian Zhang, Guangpo Tian, Fenghua Guo, Pengfei Wang

{"title":"TFP-mixer: A lightweight time and frequency combining model for multivariate long-term time series forecasting","authors":"Zhaodian Zhang, Guangpo Tian, Fenghua Guo, Pengfei Wang","doi":"10.1007/s10489-025-06562-7","DOIUrl":"10.1007/s10489-025-06562-7","url":null,"abstract":"<div><p>Time series are widely present in various fields such as financial investment, energy consumption, electricity usage, and traffic flow. By analyzing time series, we can predict future trends and patterns, which helps in making strategic decisions, optimizing resource allocation, and improving overall efficiency. Recently, most methods prioritize prediction accuracy, often overlooking memory and computational costs, which limit applicability in scenarios requiring rapid response times or high computational resources. Even when focusing solely on prediction accuracy, these methods often overlook important considerations, such as the interactions between time and frequency features, among channels, and within patches. To address these issues, we designed a lightweight time series forecasting model called TFP-Mixer, which integrates both time domain and frequency domain information. In the time domain, TFP-Mixer captures the dynamic changes and dependencies of time series through Time/Frequency interaction, Channel interaction, and Patch interaction. By using Discrete Fourier transform (DFT) to convert time series into frequency domain data, the model extracts and interacts with frequency domain features, enhancing its ability to capture frequency domain characteristics. Extensive experiments on nine real-world time series datasets show that TFP-Mixer achieves a 6.17% and 7.15% improvement over state-of-the-art (SOTA) methods. The code is available at https://github.com/SDUYanDong/TFP-Mixer</p></div>","PeriodicalId":8041,"journal":{"name":"Applied Intelligence","volume":"55 10","pages":""},"PeriodicalIF":3.4,"publicationDate":"2025-04-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143879696","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Anchor-based incomplete multi-view clustering with graph convolution network 基于锚点的不完全多视图图卷积网络聚类

IF 3.4 2区计算机科学

Applied Intelligence Pub Date : 2025-04-28 DOI: 10.1007/s10489-025-06580-5

Ao Li, Tianyu Gao, Yanbing Wang, Cong Feng

{"title":"Anchor-based incomplete multi-view clustering with graph convolution network","authors":"Ao Li, Tianyu Gao, Yanbing Wang, Cong Feng","doi":"10.1007/s10489-025-06580-5","DOIUrl":"10.1007/s10489-025-06580-5","url":null,"abstract":"<div><p>Anchor-based method has proved to be effective in recent incomplete multi-view clustering literature. Although existing methods have achieved significant success in various fields (e.g., digital treatment), they still have several limitations: (1) The construction of anchor graph insufficiently considers the graph structural information inherent in the original data. (2) Most studies are unable to sufficiently explore the correlation between the non-linear structures of representation space and the original space. In this paper, we propose a Anchor-based Incomplete Multi-view Clustering with Graph Convolution Network (AIMCG) method to address the above issues. Specifically, we first adopt graph convolution networks to extract graph information from multi-view data, and employ manifold regularization to constrain the generation of common graph representation. Subsequently, we employ an anchor-based data reconstruction method to generate anchor g raph, combining previous graph information into this process to further enhance the clustering capability. Finally, spectral clustering is applied to the anchor graph to obtain the clustering results. Experiments on 9 benchmark datasets compared with 13 advanced baselines verify the effectiveness of our AIMCG method on incomplete multi-view data.</p></div>","PeriodicalId":8041,"journal":{"name":"Applied Intelligence","volume":"55 10","pages":""},"PeriodicalIF":3.4,"publicationDate":"2025-04-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143879586","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

EPDNet: Light-weight small target detection algorithm based on pruning and logical distillation EPDNet：基于剪枝和逻辑蒸馏的轻量级小目标检测算法

IF 3.4 2区计算机科学

Applied Intelligence Pub Date : 2025-04-28 DOI: 10.1007/s10489-025-06582-3

Gaofeng Zhu, Zhixue Wang, Fenghua Zhu, Gang Xiong, Zheng Li

{"title":"EPDNet: Light-weight small target detection algorithm based on pruning and logical distillation","authors":"Gaofeng Zhu, Zhixue Wang, Fenghua Zhu, Gang Xiong, Zheng Li","doi":"10.1007/s10489-025-06582-3","DOIUrl":"10.1007/s10489-025-06582-3","url":null,"abstract":"<div><p>Drone detection technology plays a crucial role in various fields. However, due to the limited computational resources of edge devices onboard drones, achieving efficient detection using large-parameter algorithms remains challenging. Small target detection in drone-based applications faces several difficulties, including the small size of targets, limited feature information, and vulnerability to environmental interference. Moreover, existing lightweight small target detection methods often compromise detection accuracy while reducing model parameters, failing to meet the dual requirements of accuracy and efficiency in drone scenarios. To address these challenges, this paper proposes EPDNet, a lightweight small target detection algorithm designed for drone applications. First, ConvNextV2 replaces the original backbone network, incorporating a fully convolutional masked autoencoder framework combined with a self-supervised learning strategy to enhance the extraction of essential low-level features. Additionally, the EC2f feature extraction module is introduced to enable interactive modeling of contextual detail features across different target scales, orientations, and shapes. Furthermore, an adaptive channel pruning scheme is designed to reduce redundant parameters and computational complexity, thereby enhancing algorithm efficiency. Finally, the detection performance of the pruned model is further optimized using knowledge distillation. Experimental results on the VisDrone2019 aerial photography dataset demonstrate that EPDNet improves detection precision (P) by 2.6%, increases mean average precision (mAP) by 3.0%, reduces the number of parameters by 29.6%, and decreases computational cost by 17.8%. These results indicate that EPDNet effectively meets the lightweight deployment requirements of drone-based applications.</p></div>","PeriodicalId":8041,"journal":{"name":"Applied Intelligence","volume":"55 10","pages":""},"PeriodicalIF":3.4,"publicationDate":"2025-04-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143879639","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Orthogonal and spherical quaternion features for weakly supervised learning with label confidence optimization 带标签置信度优化的弱监督学习的正交和球面四元数特征

IF 3.4 2区计算机科学

Applied Intelligence Pub Date : 2025-04-28 DOI: 10.1007/s10489-025-06575-2

Heng Zhou, Ping Zhong

{"title":"Orthogonal and spherical quaternion features for weakly supervised learning with label confidence optimization","authors":"Heng Zhou, Ping Zhong","doi":"10.1007/s10489-025-06575-2","DOIUrl":"10.1007/s10489-025-06575-2","url":null,"abstract":"<div><p>Weakly supervised learning (WSL) addresses the challenge of incomplete or noisy labels, but current methods often fail to capture the complexities introduced by weak labels in feature extraction, revealing the limitations of neural networks in modeling the intricate relationships between features and labels. To address these issues, we introduce the Orthogonal and Spherical Quaternion Neural Network (OSQNN), which utilizes quaternion feature embedding with an orthogonal constraint to map real-valued features into quaternion space. This approach improves the understanding of feature-label relationships by overcoming the challenge of embedding real-world data into quaternion spaces. OSQNN maps quaternion features onto a sphere and estimates label reliability through nearest neighbors, maintaining a coherent geometric structure in feature distributions. Furthermore, quaternion convolutions are transformed into parallel grouped real-valued convolutions, enhancing processing efficiency without sacrificing the benefits of quaternion-based computations. Additionally, we propose the Label Confidence Guided Expectation-Maximization (LCGEM) algorithm, integrated into OSQNN, to more effectively capture the complex relationships between weak labels and feature distributions. Experimental results across eight datasets demonstrate the superiority of OSQNN. For instance, in SSL on CIFAR10 (20% labeled data) and CIFAR100, it achieved 91.06% and 69.16% accuracy respectively; in NSL with 40% incorrect labels on CIFAR10 and CIFAR100, the accuracies were 80.84% and 51.98%, showing its high accuracy and robustness. The ablation study highlights the role of the orthogonal constraint and spherical feature mapping in improving performance, while t-SNE visualization confirms the ability of OSQNN to learn discriminative feature representations.</p></div>","PeriodicalId":8041,"journal":{"name":"Applied Intelligence","volume":"55 10","pages":""},"PeriodicalIF":3.4,"publicationDate":"2025-04-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143879695","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0