检索结果-内蒙古大学图书馆

few-shot class-incremental learning for classification and Object Detection: A Survey

IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE 2025年第4期47卷 2924-2945页

作者： Zhang, Jinghua Liu, Li Silven, Olli Pietikainen, Matti Hu, Dewen Natl Univ Def Technol NUDT Coll Intelligence Sci & Technol Changsha 410073 Peoples R China Univ Oulu Ctr Machine Vis & Signal Anal CMVS Oulu 90570 Finland Natl Univ Def Technol NUDT Coll Elect Sci & Technol Changsha 410073 Peoples R China Univ Oulu CMVS Oulu 90570 Finland

few-shot class-incremental learning (FSCIL) presents a unique challenge in Machine learning (ML), as it necessitates the incremental learning (IL) of new classes from sparsely labeled training samples without forgetting previous knowledge. While this field has seen recent progress, it remains an active exploration area. This paper aims to provide a comprehensive and systematic review of FSCIL. In our in-depth examination, we delve into various facets of FSCIL, encompassing the problem definition, the discussion of the primary challenges of unreliable empirical risk minimization and the stability-plasticity dilemma, general schemes, and relevant problems of IL and few-shot learning (FSL). Besides, we offer an overview of benchmark datasets and evaluation metrics. Furthermore, we introduce the few-shot class-incremental classification (FSCIC) methods from data-based, structure-based, and optimization-based approaches and the few-shot class-incremental Object Detection (FSCIOD) methods from anchor-free and anchor-based approaches. Beyond these, we present several promising research directions within FSCIL that merit further investigation.

关键词： incremental learning continual learning lifelong learning lifelong learning class-incremental learning class-incremental learning catastrophic forgetting catastrophic forgetting few- shot learning few- shot learning few-shot class-incremental learning deep learning deep learning image classification image classification image classification

来源：评论

学校读者我要写书评

暂无评论

few-shot class-incremental learning for Medical Time Series classification

引用

IEEE JOURNAL OF BIOMEDICAL AND HEALTH INFORMATICS 2024年第4期28卷 1872-1882页

作者： Sun, Le Zhang, Mingyang Wang, Benyou Tiwari, Prayag Nanjing Univ Informat Sci & Technol Engn Res Ctr Digital Forens Minist Educ Nanjing 210044 Peoples R China Nanjing Univ Informat Sci & Technol Dept Jiangsu Collaborat Innovat Ctr Atmospher Envi Nanjing 210044 Peoples R China Chinese Univ Hong Kong SRIBD & SDS Shenzhen 518172 Peoples R China Halmstad Univ Sch Informat Technol Halmstad Sweden

Continuously analyzing medical time series as new classes emerge is meaningful for health monitoring and medical decision-making. few-shot class-incremental learning (FSCIL) explores the classification of few-shot new classes without forgetting old classes. However, little of the existing research on FSCIL focuses on medical time series classification, which is more challenging to learn due to its large intra-class variability. In this paper, we propose a framework, the Meta self-Attention Prototype Incrementer (MAPIC) to address these problems. MAPIC contains three main modules: an embedding encoder for feature extraction, a prototype enhancement module for increasing inter-class variation, and a distance-based classifier for reducing intra-class variation. To mitigate catastrophic forgetting, MAPIC adopts a parameter protection strategy in which the parameters of the embedding encoder module are frozen at incremental stages after being trained in the base stage. The prototype enhancement module is proposed to enhance the expressiveness of prototypes by perceiving inter-class relations using a self-attention mechanism. We design a composite loss function containing the sample classification loss, the prototype non-overlapping loss, and the knowledge distillation loss, which work together to reduce intra-class variations and resist catastrophic forgetting. Experimental results on three different time series datasets show that MAPIC significantly outperforms state-of-the-art approaches by 27.99%, 18.4%, and 3.95%, respectively.

关键词： Health monitoring medical decision-making few-shot class-incremental learning medical time series classification

来源：评论

学校读者我要写书评

暂无评论

few-shot class-incremental learning for Network Intrusion Detection Systems

IEEE OPEN JOURNAL OF THE COMMUNICATIONS SOCIETY

引用

IEEE OPEN JOURNAL OF THE COMMUNICATIONS SOCIETY 2024年 5卷 6736-6757页

作者： Di Monda, Davide Montieri, Antonio Persico, Valerio Voria, Pasquale De Ieso, Matteo Pescape, Antonio Univ Napoli Federico II Dipartimento Ingn Elettr & Tecnol Informaz I-80125 Naples Italy IMT Sch Adv Studies I-55100 Lucca Italy

In today's digital landscape, critical services are increasingly dependent on network connectivity, thus cybersecurity has become paramount. Indeed, the constant escalation of cyberattacks, including zero-day exploits, poses a significant threat. While Network Intrusion Detection Systems (NIDSs) leveraging machine-learning and deep-learning models have proven effective in recent studies, they encounter limitations such as the need for abundant samples of malicious traffic and full retraining upon encountering new attacks. These limitations hinder their adaptability in real-world scenarios. To address these challenges, we design a novel NIDS capable of promptly adapting to classify new attacks and provide timely predictions. Our proposal for attack-traffic classification adopts few-shot class-incremental learning (FSCIL) and is based on the Rethinking few-shot (RFS) approach, which we experimentally prove to overcome other FSCIL state-of-the-art alternatives based on either meta-learning or transfer learning. We evaluate the proposed NIDS across a wide array of cyberattacks whose traffic is collected in recent publicly available datasets to demonstrate its robustness across diverse network-attack scenarios, including malicious activities in an Internet-of-Things context and cyberattacks targeting servers. We validate various design choices as well, involving the number of traffic samples per attack available, the impact of the features used to represent the traffic objects, and the time to deliver the classification verdict. Experimental results witness that our proposed NIDS effectively retains previously acquired knowledge (with over 94% F1-score) while adapting to new attacks with only few samples available (with over 98% F1-score). Thus, it outperforms non-FSCIL state of the art in terms of classification effectiveness and adaptation time. Moreover, our NIDS exhibits high performance even with traffic collected within short time frames, achieving 95% F1-score whi

关键词： Power capacitors Training Metalearning Computer crime Denial-of-service attack Transfer learning Network intrusion detection Telecommunication traffic Vectors Traffic control Attack-traffic classification deep learning few-shot class-incremental learning network intrusion detection system network security

来源：评论

学校读者我要写书评

暂无评论

few-shot class-incremental learning by Sampling Multi-Phase Tasks

引用

IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE 2023年第11期45卷 12816-12831页

作者： Zhou, Da-Wei Ye, Han-Jia Ma, Liang Xie, Di Pu, Shiliang Zhan, De-Chuan Nanjing Univ State Key Lab Novel Software Technol Nanjing 210023 Peoples R China Hikvis Res Inst Hangzhou 310051 Peoples R China

New classes arise frequently in our ever-changing world, e.g., emerging topics in social media and new types of products in e-commerce. A model should recognize new classes and meanwhile maintain discriminability over old classes. Under severe circumstances, only limited novel instances are available to incrementally update the model. The task of recognizing few-shot new classes without forgetting old classes is called few-shot class-incremental learning (FSCIL). In this work, we propose a new paradigm for FSCIL based on meta-learning by learning Multi-phase incremental Tasks (Limit), which synthesizes fake FSCIL tasks from the base dataset. The data format of fake tasks is consistent with the 'real' incremental tasks, and we can build a generalizable feature space for the unseen tasks through meta-learning. Besides, Limit also constructs a calibration module based on transformer, which calibrates the old class classifiers and new class prototypes into the same scale and fills in the semantic gap. The calibration module also adaptively contextualizes the instance-specific embedding with a set-to-set function. Limit efficiently adapts to new classes and meanwhile resists forgetting over old classes. Experiments on three benchmark datasets (CIFAR100, miniImageNet, and CUB200) and large-scale dataset, i.e., ImageNet ILSVRC2012 validate that Limit achieves state-of-the-art performance.

关键词： Task analysis Training Power capacitors Adaptation models Data models Calibration Transformers few-shot class-incremental learning transformer meta-learning class-incremental learning

来源：评论

学校读者我要写书评

暂无评论

few-shot class-incremental learning via Cross-Modal Alignment with Feature Replay 7th

Few-Shot Class-Incremental Learning via Cross-Modal Alignmen...

引用

7th Chinese Conference on Pattern Recognition and Computer Vision

作者： Li, Yanan He, Linpu Lin, Feng Wang, Donghui Zhejiang Lab Res Ctr Frontier Fundamental Studies Hangzhou Peoples R China Zhejiang Univ Dept Comp Sci & Technol Hangzhou Peoples R China

ISBN: (纸本)9789819784868;9789819784875

few-shot class-incremental learning (FSCIL) studies the problem of continually learning novel concepts from a limited training data without catastrophically forgetting the old ones at the meantime. While most existing works are established on the premise of learning from scratch, growing efforts have been devoted to incorporating the benefits of pre-trained Vision-Language Models (VLMs) within the FSCIL solution, considering that these models have shown powerful generalization abilities in zero-shot/few-shot learning. In this paper, we propose a simple yet effective FSCIL framework that well leverages the prior knowledge of the CLIP model to attack the stability-plasticity dilemma. Considering the semantic gap between the pre-trained and downstream data, we first combine soft prompts with visual adaptation to effectively accommodate the prior knowledge from both branches. Then, we condition the textual prototype on each visual input to adaptively capture the instance-specific information, taking account of their intrinsic heterogeneous structures. On top of this framework, we employ a simple feature replay strategy that models each class as a Gaussian distribution to alleviate the task interference in each new session. Extensive experimental results on three benchmarks, i.e., CIFAR100, CUB200 and miniImageNet, show that our proposed method can achieve compelling FSCIL results.

关键词： few-shot class-incremental learning Soft Prompt learning Visual Adapter Cross-Modal Adapter

来源：评论

学校读者我要写书评

暂无评论

few-shot class-incremental learning via Compact and Separable Features for Fine-Grained Vehicle Recognition

引用

IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS 2022年第11期23卷 21418-21429页

作者： Li, De-Wang Huang, Hua Beijing Inst Technol Sch Comp Sci & Technol Beijing 100081 Peoples R China Beijing Normal Univ Sch Artificial Intelligence Beijing 100875 Peoples R China

Most of the existing deep learning-based fine-grained vehicle recognition methods collect a large-scale training set in advance and train a model based on the closed-world assumption. However, in the real world, new classes of vehicles are released over time but it is difficult to collect sufficient labeled data for new classes, which results in a typical few-shot class-incremental learning problem (FSCIL). To solve this problem, this work proposes a compact and separable feature learning method (CSFL) which exploits a decoupled learning scheme to prevent the feature extractor from updating during class-incremental learning. CSFL trains an initial model to learn discriminative features of fine-grained vehicles using deep metric learning. Then an incremental linear discriminant analysis algorithm is applied to the learned features to further discriminate potential confused classes. Specifically, the decoupled components share the same objective of enhancing intra-class compactness and inter-class separability, which is beneficial for classification. Extensive experiments on three fine-grained vehicle datasets demonstrate that the proposed CSFL achieves better results than state-of-the-art incremental learning methods, validating the importance of compact and separable features in the problem of FSCIL.

关键词： Feature extraction Training Power capacitors Adaptation models Task analysis Image recognition Measurement Fine-grained vehicle recognition few-shot class-incremental learning deep metric learning linear discriminant analysis

来源：评论

学校读者我要写书评

暂无评论

few-shot class-incremental learning for EEG-Based Emotion Recognition 29th

Few-Shot Class-Incremental Learning for EEG-Based Emotion Re...

引用

29th International Conference on Neural Information Processing

作者： Ma, Tian-Fang Zheng, Wei-Long Lu, Bao-Liang Shanghai Jiao Tong Univ Dept Comp Sci & Engn 800 Dongchuan Rd Shanghai 200240 Peoples R China Shanghai Jiao Tong Univ RuiJin Hosp RuiJin Mihoyo Lab Sch Med 197 Ruijin 2nd Rd Shanghai 200020 Peoples R China Shanghai Jiao Tong Univ Key Lab Shanghai Commiss Intelligent Interact & C 800 Dongchuan Rd Shanghai 200240 Peoples R China Shanghai Jiao Tong Univ RuiJin Hosp Clin Neurosci Ctr Sch Med 197 Ruijin 2nd Rd Shanghai 200020 Peoples R China Shanghai Jiao Tong Univ Brain Sci & Technol Res Ctr 800 Dong Chuan Rd Shanghai 200240 Peoples R China

ISBN: (纸本)9789819916412;9789819916429

Current advanced deep neural networks can greatly improve the performance of emotion recognition tasks in affective Brain-Computer Interfaces (aBCI). Basic human emotions could be induced and electroencephalographic (EEG) signals could be simultaneously recorded. While data of basic common emotions are easier to collect, some complex emotions are low resource in terms of data size and label quality in real life, which would limit the utility of EEG-based emotion recognition models. To enhance the model adaptive capacity of new emotions with few samples, we introduce a few-shot class-incremental deep learning model for emotion recognition. The proposed model consists of a graph convolutional networks (GCN) and a linear classifier. By training the whole network on a base set in a preliminary stage, and fine-tuning the parameters of the linear classifier with very few shots of labeled samples, the model can incrementally learn new types of emotions while preserving knowledge of the old ones. Our experimental results on the SEED-V dataset show that even with very limited new class samples, the fine-tuned pre-trained model could have a fairly good performance on the test set with more emotion classes.

关键词： EEG Deep learning few-shot class-incremental learning

来源：评论

学校读者我要写书评

暂无评论

FSCIL-EACA: few-shot class-incremental learning Network Based on Embedding Augmentation and classifier Adaptation for Image classification

引用

Chinese Journal of Electronics 2024年第1期33卷 139-152页

作者： Ruru ZHANG Haihong E Meina SONG School of Computer Science Beijing University of Posts and Telecommunications Education Department Information Network Engineering Research Center Beijing University of Posts and Telecommunications

The ability to learn incrementally is critical to the long-term operation of AI systems. Benefiting from the power of few-shot class-incremental learning(FSCIL), deep learning models can continuously recognize new classes with only a few samples. The difficulty is that limited instances of new classes will lead to overfitting and exacerbate the catastrophic forgetting of the old classes. Most previous works alleviate the above problems by imposing strong constraints on the model structure or parameters, but ignoring embedding network transferability and classifier adaptation(CA), failing to guarantee the efficient utilization of visual features and establishing relationships between old and new classes. In this paper, we propose a simple and novel approach from two perspectives: embedding bias and classifier bias. The method learns an embedding augmented(EA) network with cross-class transfer and class-specific discriminative abilities based on self-supervised learning and modulated attention to alleviate embedding bias. Based on the adaptive incremental classifier learning scheme to realize incremental learning capability,guiding the adaptive update of prototypes and feature embeddings to alleviate classifier bias. We conduct extensive experiments on two popular natural image datasets and two medical datasets. The experiments show that our method is significantly better than the baseline and achieves state-of-the-art results.

关键词： few-shot class-incremental learning Embedding augmentation classifier adaptation Image classification

来源：评论

学校读者我要写书评

暂无评论

Prompt-Based Concept learning for few-shot class-incremental learning

引用

IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS FOR VIDEO TECHNOLOGY 2025年第5期35卷 4991-5005页

作者： Li, Shuo Liu, Fang Jiao, Licheng Li, Lingling Chen, Puhua Liu, Xu Ma, Wenping Xidian Univ Int Res Ctr Intelligent Percept & Computat Sch Artificial Intelligence Key Lab Intelligent Percept & Image Understanding Xian 710071 Peoples R China

few-shot class-incremental learning (FSCIL) faces a huge stability-plasticity challenge due to continuously learning knowledge from new classes with a small number of training samples without forgetting the knowledge of previously seen old classes. To alleviate this challenge, we propose a novel method called Prompt-based Concept learning (PCL) for FSCIL, which generalizes conceptual knowledge learned from old classes to new classes by simulating human learning capabilities. In our PCL, in the base session, we simultaneously learn common basic concepts from the training data and the class-concept weight of each class in a prompt learning manner, and in each incremental session, class-concept weights between new classes and previously learned basic concepts are learned to achieve incremental learning. Furthermore, in order to avoid catastrophic forgetting, we propose a distribution estimation module to retain feature distributions of previously seen classes and a data replay module to randomly sample features of previously seen classes in incremental sessions. We verify the effectiveness of our PCL on widely used benchmarks, such as miniImageNet, CIFAR-100, and CUB-200. Experimental results show that our PCL achieves competitive results compared with other state-of-the-art methods, especially we achieve an average accuracy of 94.02% across all sessions on the miniImageNet benchmark.

关键词： Power capacitors Adaptation models Circuits and systems Birds Training data Stability plasticity incremental learning Estimation Training Reviews Concept learning prompt learning class-incremental learning few-shot class-incremental learning

来源：评论

学校读者我要写书评

暂无评论

A prompt regularization approach to enhance few-shot class-incremental learning with Two-Stage classifier

引用

NEURAL NETWORKS 2025年 188卷 107453页

作者： Hao, Meilan Gu, Yizhan Dong, Kejian Tiwari, Prayag Lv, Xiaoqing Ning, Xin Hebei Univ Engn Sch Informat & Elect Engn Handan 056038 Peoples R China Chinese Acad Sci Inst Semicond Beijing 100083 Peoples R China Halmstad Univ Sch Informat Technol SE-30118 Halmstad Sweden Beijing Ratu Technol Co Ltd Beijing 100096 Peoples R China

With a limited number of labeled samples, few-shot class-incremental learning (FSCIL) seeks to efficiently train and update models without forgetting previously learned tasks. Because pre-trained models can learn extensive feature representations from big existing datasets, they offer strong knowledge foundations and transferability, which makes them useful in both few-shot and incremental learning scenarios. Additionally, Prompt learning improves pre-trained deep learning models' performance on downstream tasks, particularly in large-scale language or vision models. In this paper, we propose a novel Prompt Regularization (PrRe) approach to maximize the fusion of prompts by embedding two different prompts, the Task Prompt and the Global Prompt, inside a pre-trained Vision Transformer (ViT). In the classification phase, we propose a Two-Stage classifier (TSC), utilizing K-Nearest Neighbors for base session and a Prototype classifier for incremental sessions, integrated with a global self-attention module. Through experiments on multiple benchmark tests, we demonstrate the effectiveness and superiority of our method. The code is available at https://***/gyzzzzzzzz/PrRe.

关键词： Prompt regularization Two-Stage classifier Global self-attention module few-shot class-incremental 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：