检索结果-内蒙古大学图书馆

few-shot learning With Enhancements to Data Augmentation and Feature Extraction

IEEE TRANSACTIONS ON NEURAL NETWORKS AND learning SYSTEMS 2025年第4期36卷 6655-6668页

作者： Zhang, Yourun Gong, Maoguo Li, Jianzhao Feng, Kaiyuan Zhang, Mingyang Xidian Univ Hangzhou Inst Technol Hangzhou 311200 Peoples R China Xidian Univ Key Lab Collaborat Intelligence Syst Minist Educ Xian 710071 Peoples R China Xidian Univ Key Lab Collaborat Intelligence Syst Minist Educ Xian 010028 Peoples R China Xidian Univ Key Lab Collaborat Intelligence Syst Minist Educ Xian 710071 Peoples R China

The few-shot image classification task is to enable a model to identify novel classes by using only a few labeled samples as references. In general, the more knowledge a model has, the more robust it is when facing novel situations. Although directly introducing large amounts of new training data to acquire more knowledge is an attractive solution, it violates the purpose of few-shot learning with respect to reducing dependence on big data. Another viable option is to enable the model to accumulate knowledge more effectively from existing data, i.e., improve the utilization of existing data. In this article, we propose a new data augmentation method called self-mixup (SM) to assemble different augmented instances of the same image, which facilitates the model to more effectively accumulate knowledge from limited training data. In addition to the utilization of data, few-shot learning faces another challenge related to feature extraction. Specifically, existing metric-based few-shot classification methods rely on comparing the extracted features of the novel classes, but the widely adopted downsampling structures in various networks can lead to feature degradation due to the violation of the sampling theorem, and the degraded features are not conducive to robust classification. To alleviate this problem, we propose a calibration-adaptive downsampling (CADS) that calibrates and utilizes the characteristics of different features, which can facilitate robust feature extraction and benefit classification. By improving data utilization and feature extraction, our method shows superior performance on four widely adopted few-shot classification datasets.

关键词： Feature extraction Training Data models Testing Manifolds Data augmentation Task analysis feature extraction few-shot learning image classification

来源：评论

学校读者我要写书评

暂无评论

few-shot learning for estimating gear wear severity towards digital twinning

引用

ENGINEERING FAILURE ANALYSIS 2025年 170卷

作者： Cohen, Roee Bachar, Lior Matania, Omri Bortman, Jacob Ben Gurion Univ Negev Dept Mech Engn BGU PHM Lab POB 653 IL-8410501 Beer Sheva Israel

Digital twins hold the potential to revolutionize gear health monitoring by leveraging dynamic modeling and AI advancements, providing real-time insights and enabling proactive maintenance. Wear is one of the most critical fault mechanisms in gears, yet vibration-based gear diagnosis has mainly progressed for localized tooth faults due to the complexities associated with distributed wear. This study proposes a novel strategy for gear wear monitoring aligning with digital twins' foundations, by integrating experiments, dynamic modeling, physical preprocessing, and machine learning. First, we propose an unsupervised learning-based sensitive health indicator for early wear detection, grounded in advanced physical preprocessing. Then, we introduce a novel few-shot learning approach for estimating wear severity, addressing the scarcity of labeled faulty measured data in real-world scenarios. We train a severity regressor using an extensive simulated training dataset, augmented and adapted to the inspected gear through transfer function estimation. The few-shot learning framework relies on measured training data, which includes healthy samples and a few faulty samples from a one label. The transfer of knowledge between the simulated and measured training datasets enables the severity regressor to make predictions on unseen measured data. The proposed strategy is demonstrated through controlled-degradation tests in gear wear, showcasing its superior performance in both fault detection and severity estimation, comparable to that of a fully supervised model, as if labeled measured data were available. Above all, this study highlights how a synergy between physicsbased and AI-based approaches enhances performance while addressing practical challenges in wear diagnosis, paving the way for the implementation of digital twins in future endeavors.

关键词： Mechanical engineering Spur gear Steel Wear Poor lubrication Vibration signals Fault severity estimation few-shot learning

来源：评论

学校读者我要写书评

暂无评论

few-shot learning Augmented with Image Transformation for Multiclass Structural Damage Classification

引用

JOURNAL OF COMPUTING IN CIVIL ENGINEERING 2025年第3期39卷

作者： Dunphy, Kyle Buwaneswaran, Madhushan Grolinger, Katarina Sadhu, Ayan Western Univ Dept Civil & Environm Engn London ON N6A 5B9 Canada Western Univ Dept Elect & Comp Engn London ON N6A 5B9 Canada Western Univ Western Acad Adv Res Dept Civil & Environm Engn London ON N6A 5B9 Canada

The application of machine learning (ML) as an apparatus for structural health monitoring (SHM) has become increasingly prevalent recently as the domain moves toward autonomous structural inspections. Although significant work has been conducted to integrate ML in SHM, many domain-specific issues adopting these technologies are still prevalent. For instance, ML is characterized as a data-intensive technique, requiring a significant number of samples to properly train a new model which are often unavailable in SHM applications. Furthermore, the generalization of these models to new categories of damages and structural and material types results in inferior damage classification. Therefore, to address the scarcity of data within SHM, few-shot learning (FSL) models, such as prototypical networks, have been recently explored as they are capable of training accurate classification models with limited images. However, the use of limited data results in model overfitting and may not adapt well to novel classes of data originating from new material and structural sources. In this paper, the effect of several image transformation techniques on the performance of a prototypical network is investigated concerning surface-level damages for concrete and asphalt structures. The effects of intramaterial data sets (data sets derived from the same material type), and intermaterial data sets (data sets derived from different material types) are investigated to understand and quantify the domain adaptation of these models. It was demonstrated that for k>2, histogram equalization, logarithmic transform, and power transform performed marginally better (1%-5% for both material scenarios) than standard grayscale images when training the chosen prototypical network. The use of phase stretch transform and histogram equalization provided a better reduction to overfitting for both material scenarios (1%-5% and 1%-3%, respectively) when compared to grayscale, further demonstrating the effectiv

关键词： few-shot learning Prototypical networks Image transformation Damage classification Concrete Asphalt

来源：评论

学校读者我要写书评

暂无评论

few-shot learning for signal detection in wideband spectrograms

引用

DIGITAL SIGNAL PROCESSING 2025年 162卷

作者： Li, Weihao Deng, Wen Wang, Keren You, Ling Huang, Zhitao Natl Univ Def Technol Coll Elect Sci & Technol Changsha 410073 Peoples R China Natl Key Lab Blind Signal Proc Chengdu 610041 Peoples R China State Key Lab Complex Electromagnet Environm Effec Luoyang 471003 Peoples R China

Signal detection in the wideband plays an important role in spectrum sensing or reconnaissance tasks. Considering the visualization benefits of the spectrogram and the great developments in deep learning object detection, an increasing number of researchers have implemented deep learning-based signal detection in wideband spectrograms, which obtained remarkable performance. Most existing detection models rely on the availability of abundant labeled training data, but for signal classes with little labeled training data, the detection performance will deteriorate significantly. In this paper, a few-shot signal detection model is proposed to solve this problem. The proposed model is pretrained on abundantly labeled base signal classes and aims to detect novel classes given only a few labeled samples. The model is built on an existing base detector designed specifically for signal detection, and a class-specific convolution kernel generator (CCKG) is proposed to generate convolution kernels by template signals for predictions of signal center frequency and shape attributes. Benefiting from a three-stage meta-learning procedure, the CCKG can play a significant role with only a few input samples. Comprehensive experiments with a simulated signal superimposed on real background dataset and a real-world dataset demonstrate that the proposed method yields significantly better performance than the well-established baseline models.

关键词： few-shot learning Signal detection Wideband spectrogram Meta-learning Deep learning

来源：评论

学校读者我要写书评

暂无评论

Predicting the productivity of fractured horizontal wells using few-shot learning

引用

Petroleum Science 2025年第2期22卷 787-804页

作者： Sen Wang Wen Ge Yu-Long Zhang Qi-Hong Feng Yong Qin Ling-Feng Yue Renatus Mahuyu Jing Zhang State Key Laboratory of Deep Oil and Gas China University of Petroleum East China School of Petroleum Engineering China University of Petroleum East China Research Institute of Exploration and Development SINOPEC Jiangsu Oilfeld Shandong Institute of Petroleum and Chemical Technology Research Institute of Petroleum Exploration and Development Petro China Exploration and Development Research Institute Petro China Xinjiang Oilfeld Company

Predicting the productivity of multistage fractured horizontal wells plays an important role in exploiting unconventional *** recent years,machine learning（ML） models have emerged as a new approach for such ***,the scarcity of sufficient real data for model training often leads to imprecise predictions,even though the models trained with real data better characterize geological and engineering *** tackle this issue,we propose an ML model that can obtain reliable results even with a small amount of data *** model integrates the synthetic minority oversampling technique（SMOTE） to expand the data volume,the support vector machine（SVM） for model training,and the particle swarm optimization（PSO） algorithm for optimizing *** enhance the model performance,we conduct feature fusion and dimensionality ***,we examine the influences of different sample sizes and ML models for *** proposed model demonstrates higher prediction accuracy and generalization ability,achieving a predicted R2value of up to 0.9 for the test set,compared to the traditional ML techniques with an R2of *** model accurately predicts the production of fractured horizontal wells even with limited samples,supplying an efficient tool for optimizing the production of unconventional ***,the model holds the potential applicability to address similar challenges in other fields constrained by scarce data samples.

关键词： Fractured horizontal well Machine learning SMOTE few-shot learning Prediction Optimization

来源：评论

学校读者我要写书评

暂无评论

Convolutional Cross-Modal Autoencoder-Based few-shot learning for Data Augmentation with Application to Alzheimer Dementia Diagnosis

引用

COGNITIVE COMPUTATION 2025年第1期17卷 1-14页

作者： Bazine, Othmane Rai, Omar Aiadi, Oussama Hedjam, Rachid Khaldi, Belal Zhong, Guoqiang Univ Kasdi Merbah Comp Sci Ouargla 30000 Algeria Artificial Intelligence & Informat Technol Lab LIN Ouargla 30000 Algeria Bishops Univ Comp Sci Dept Sherbrooke PQ Canada Ocean Univ China Comp Sci & Technol 238 Songling Rd Qingdao 266100 Peoples R China

This paper presents a novel deep few-shot learning method for magnetic resonance images-based Alzheimer's dementia (AD) diagnosis. The proposed method consists of two main phases namely data augmentation and data classification. With regard to data augmentation and, to deal with data scarcity issues, we designed a convolutional cross-modal autoencoder (CCMAE) model for data generation. This model, which consists of two encoders and one decoder, receives two image modalities namely longitudinal and cross-section MRI, and generates a new cross-section image. We opt for a convolutional version of the autoencoder to capture the spatial information more effectively and reduce the number of trainable parameters. Moreover, to make the model able to perform deep analysis of input image, we establish a skip connection strategy between the first encoder and the decoder similar to the UNet mechanism. With regard to classification, we design a convolutional neural network-based model in which both textual and visual features are fused to strengthen the network performance and produce more reliable decisions. A comprehensive experiment on a publicly available dataset has been conducted to demonstrate the effectiveness of the proposed method compared to some related works. The code is publicly available at: https://***/Bazine-Othmane/scientific-paper-code.

关键词： few-shot learning Convolutional autoencoder Convolutional neural networks Dementia Alzheimer Magnetic imaging resonance

来源：评论

学校读者我要写书评

暂无评论

Boosting meta-training with base class information for robust few-shot learning

引用

ENGINEERING APPLICATIONS OF ARTIFICIAL INTELLIGENCE 2025年 152卷

作者： Jiang, Weihao Liu, Guodong He, Di He, Kun Huazhong Univ Sci & Technol Sch Comp Sci & Technol Wuhan 430000 Peoples R China Peking Univ Sch Intelligence Sci & Technol Beijing 100004 Peoples R China

few-shot learning aims to develop classifiers that can recognize new, unseen classes with only a few labeled examples. Meta-learning, with methods like Model-Agnostic Meta-learning (MAML) and Prototypical Networks, has emerged as a key approach for this challenge. A recent advancement, Meta-Baseline, utilizes sequential pre-training and meta-training to achieve state-of-the-art performance. However, during meta-training, designed to further adapt the model to few-shot classification tasks, the class transferability gained during pre-training can be compromised, leading to suboptimal performance. We propose an end-to-end training paradigm with two alternating loops. The outer loop calculates cross-entropy loss across the entire training set while updating only the final linear layer. The inner loop uses the original meta-learning method to calculate the loss and integrate gradients from the outer loss to guide parameter updates. This approach enables effective adaptation to few-shot learning tasks while preserving robust generalization, outperforming existing baselines. Our findings suggest that leveraging information from the entire training set and the meta-learning training paradigm could mutually enhance one another. Our extensive experiments on the miniImageNet, tieredImageNet, and CUB datasets demonstrate the effectiveness of our method. In the 5-way 1-shot setting, we achieve accuracies of 64.01%, 69.73%, and 68.07%, outperforming the best baselines by 0.94%, 1.11%, and 1.94%, respectively. In the 5-way 5-shot setting, our accuracies are 81.00%, 84.91%, and 83.44%, outperforming the best baselines by 1.74%, 1.17%, and 3.37%, respectively. Additionally, our framework is model-agnostic, yielding significant performance improvements when integrated with existing baseline training frameworks, providing an approximate 1% performance boost.

关键词： few-shot learning Meta-learning Base class information End-to-end

来源：评论

学校读者我要写书评

暂无评论

Cross-Domain PolSAR Image Classification Using Complex-Valued few-shot learning Network

引用

IEEE TRANSACTIONS ON AEROSPACE AND ELECTRONIC SYSTEMS 2025年第2期61卷 2450-2465页

作者： Cao, Yice Wu, Zhenhua Chen, Jie Huang, Zhixiang Yang, Lixia Anhui Univ Sch Elect & Informat Engn Key Lab Intelligent Comp & Signal Proc Minist Educ Hefei 230601 Peoples R China China Elect Technol Grp Corp Res Inst 38 Hefei 230088 Peoples R China

With the impetus of supervised deep learning-based classifiers requiring extensive human-annotated training sets, polarimetric synthetic aperture (PolSAR) image classification has advanced remarkably. However, obtaining dependable labeled samples is labor-intensive and time-consuming. Furthermore, due to actual domain disparities, noticeable performance degradation often occurs when applying a trained classifier to unseen domains. Therefore, this article proposes a complex-valued cross-domain few-shot learning classification (CCFSLC) method for PolSAR images. The goal is to integrate domain adaptation with few-shot learning under the complex-valued network framework to remedy the aforementioned problems. For the proposed CCFSLC, based on the source domain with adequate label information, the created transferable knowledge learning module is first trained to learn an effective complex-valued feature encoder (CVFE) for extracting discriminative transferable knowledge. Subsequently, the deep few-shot learning module, constructed using the pretrained CVFE, undergoes training episodes in both source and target domains to learn reliable domain-invariant features, utilizing just minimal target labeled samples. Meanwhile, the adversarial domain adaptation module is employed to eliminate domain shift, thereby further enhancing cross-domain classification accuracy. The proposed CCFSLC mainly focuses on reducing the domain gap to recognize new categories in unseen domains with only a few annotated samples, while exploring more comprehensive and abundant discriminative information without compromising the integrity of the raw PolSAR data. Comprehensive experimental results on typical datasets demonstrate the superiority of CCFSLC over state-of-the-art methods for cross-domain PolSAR image classification.

关键词： few shot learning Image classification Feature extraction Training Adaptation models Data mining Aerospace and electronic systems Scattering Accuracy Testing Complex-valued neural network cross-domain polarimetric synthetic aperture (PolSAR) image classification domain adaptation few-shot learning

来源：评论

学校读者我要写书评

暂无评论

Distribution-Agnostic Probabilistic few-shot learning for Multimodal Recognition and Prediction

引用

IEEE TRANSACTIONS ON AUTOMATION SCIENCE AND ENGINEERING 2025年 22卷 1941-1957页

作者： Wang, Di Xian, Xiaochen Li, Haidong Wang, Dong Shanghai Jiao Tong Univ Sch Mech Engn Dept Ind Engn & Management Shanghai 200240 Peoples R China Univ Florida Dept Ind & Syst Engn Gainesville FL 32611 USA Univ Chinese Acad Sci Sch Econ & Management Dept Management Sci Beijing 100190 Peoples R China

In industrial scenarios with insufficient sensor data, intelligent few-shot failure mode recognition and remaining useful lifetime (RUL) prediction are critically essential for effective prognostics and health management. Existing few-shot learning (FSL) methods focus on either the failure mode recognition as a classification problem or the RUL prediction as a regression problem, failing to capture the dependence between failure modes and RUL given that units under different failure modes present distinct degradation characteristics. To address the issue, this paper proposes a distribution-agnostic probabilistic FSL method for multimodal recognition and prediction of operating units. The proposed model establishes a neural network with prototypes to solve a few-shot classification- and regression-integrated problem. To fully capture the uncertainty caused by limited sensor data, we develop multimodal Bayesian model-agnostic meta-learning (MBMAML) for the probabilistic modeling of failure modes and the RUL under multiple failure modes. We construct the loss function based on probabilistic modeling that captures the interaction between failure modes and RUL for model training. Finally, the proposed model adaptively learns the approximate distributions of failure modes and RUL for a new operating unit. We evaluate the proposed model performance through a case study on the degradation of aircraft gas turbine engines. Note to Practitioners-Failure mode recognition and RUL prediction are essential in prognostics health management (PHM) to avoid unexpected failures of units in industrial systems, such as aircraft gas turbine engines. However, insufficient sensor data are quite common issue in industrial scenarios due to expensive sensor deployment, the difficulty of installing sensors to certain special mechanical equipment, and so on. This paper aims to develop a FSL method to jointly recognize the failure mode and predict the RUL of a unit based on insufficient sensor da

关键词： few-shot learning classification- and regression-integrated problem probabilistic modeling MBMAML failure mode recognition RUL prediction

来源：评论

学校读者我要写书评

暂无评论

A cognitive few-shot learning for medical diagnosis: A case study on cleft lip and palate and Parkinson's disease

引用

EXPERT SYSTEMS WITH APPLICATIONS 2025年 262卷

作者： Yin, Pei Song, Junjie Bouteraa, Yassine Qian, Leren Martin, Diego Khishe, Mohammad Univ Shanghai Sci & Technol Business Sch Shanghai 200093 Peoples R China Univ Shanghai Sci & Technol Sch Intelligent Emergency Management Shanghai 200093 Peoples R China Prince Sattam Bin Abdulaziz Univ Coll Comp Engn & Sci Dept Comp Engn Al Kharj 11942 Saudi Arabia Arizona State Univ Sch Comp & Augmented Intelligence Tempe AZ 85281 USA Univ Valladolid Escuela Ingn Informet Segovia Dept Comp Sci Segovia 40005 Spain Yuan Ze Univ Innovat Ctr Artificial Intelligence Applicat Taoyuan Taiwan Jadara Univ Res Ctr Irbid Jordan

This paper presents a novel cognitive few-shot learning (CFSL) for the diagnosis of cleft lip and palate and Parkinson's diseases. The proposed method utilizes computational analysis of paralinguistic features to expedite the diagnostic process. Unlike other methods that rely on complex and fragmented representations, CFSL trains itself to recognize patterns that are easily interpretable by humans. Rather than learning a single, unstructured metric space, CFSL combines the outputs of individual landmark (LM) learners by mapping LMs into semi- formation spaces. In order to assess the effectiveness of CFSL, we conducted a comparative analysis with seven distinct FSL-based models, including momentum contrastive learning for FSL (MCFSL), self-updating FSL (SUFSL), mutual info multi-attention FSL (MAMIFSL), dual class representation FSL (DCRFSL), Improved FSL (IFSL), meta-knowledge for FSL (MKFSL), and prototypical networks (ProtoNet), using three popular datasets, namely GPRS, CIEMPIESS, and PC-GITA. The findings indicate that CFSL demonstrated superior performance compared to the highest-performing baseline frameworks for the 5-shot (5-sh) and 1-shot (1-sh), having an average enhancement of 4.39% and 4.49%, respectively. CFSL demonstrated better performance than the ProtoNet baseline in both 1-sh and 5-sh across all datasets, with an improvement of 12.966% and 11.033%, respectively. In addition, we performed ablation tests to assess the effects of variables such as the density of LMs, the structure of the network, the distance measure used, and the positioning of LMs. The CFSL approach, if adopted in hospitals, has the potential to enhance the precision and efficiency of diagnosis for cleft lip and palate as well as Parkinson's disease.

关键词： Cognitive few-shot learning approach Neurodegenerative disorder few-shot learning Medical diagnosis

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：