电子学报2025，Vol.53Issue(6)：1713-1740,28.DOI:10.12263/DZXB.20240750
连续学习方法与其在视觉任务中的应用
Continual Learning Methods and Applications in Computer Vision
摘要
Abstract
Continual learning generally refers to the ability of intelligent algorithms and agents to learn and adapt to a dynamic and changing world,enabling these algorithms to continually acquire,update,accumulate,and utilize knowledge throughout their deployment cycle.Continual learning technologies endow intelligent systems with the prospects and capa-bilities of adaptive development.In the context of deep learning,continual learning specifically refers to the ability to learn from non-stationary data streams and adapt to changing training objectives.This task often faces the challenge of catastroph-ic forgetting,where learning new tasks typically results in a significant decline in performance on previously learned tasks.In recent years,with the rapid development of deep learning in various fields such as language and vision,numerous ad-vancements have emerged,effectively extending the understanding and application of continual learning.This work con-ducts a relatively extensive and comprehensive survey of existing continual learning research,analyzing it from multiple perspectives including fundamental definition,representative methods,applications in the visual domain.Finally,this paper also discusses the current cutting-edge developments and future research trends in continual learning.Based on the discus-sion of relevant work in the field of continual learning,we hope this review can effectively promote the development and ex-ploration of this field and subsequent research endeavors.关键词
连续学习/持续学习/计算机视觉/通用人工智能(AI)/深度学习/神经网络Key words
continual learning/incremental learning/computer vision/general artificial intelligence/deep learning/neural networks分类
信息技术与安全科学引用本文复制引用
方岩,魏云超,丛润民,左旺孟,赵耀..连续学习方法与其在视觉任务中的应用[J].电子学报,2025,53(6):1713-1740,28.基金项目
科技创新2030-新一代人工智能重大专项(No.2021ZD0112100) Technological Innovation 2030:New Generation Artificial Intelligence Major Project(No.2021ZD0112100) （No.2021ZD0112100）
