A deep reinforcement active learning method for multi-label image classification

doi:10.1016/j.cviu.2025.104351

2025-06-01

PRE

DOI：10.1016/j.cviu.2025.104351

原文链接

原文求助

摘要

En 中文

主动学习是一种广泛应用于解决深度学习模型中样本标注高成本的方法，并在近年来取得了显著成功。然而，大多数现有的主动学习方法仅关注单标签图像分类，在多标签图像的背景下应用有限。为解决此问题，我们提出了一种基于强化学习策略的新型多标签主动学习方法。所提出的方法引入了一种考虑多标签图像预期误差减少的强化主动学习框架，使其适用于多标签分类模型。此外，我们开发了一个多标签强化主动学习模块（MLRAL），该模块采用演员-评论家策略和近端策略优化算法（PPO）。我们的状态和奖励函数考虑了多标签相关性，以准确评估未标记样本对当前模型状态的潜在影响。我们在包括VOC 2007、MS-COCO、NUS-WIDE和ODIR在内的多种多标签图像分类任务上进行了实验。我们还与多种分类模型进行了比较，实验结果表明，我们的方法在各种任务上优于现有方法，证明了所提出方法的优越性和有效性。

Keyword:

Active learning

Multi-label image classification

Deep learning

Reinforcement learning

期刊

Computer Vision and Image Understanding

IF:

3.5

论文数:

555

被引数:

7.3K

机构

Southwestern University of Finance and Economics

学者数:

938

论文数: 584

被引数: 56

University of Electronic Science and Technology of China

学者数:

5.5K

论文数: 2.2K

被引数: 4.0W

引用论文

Learning Loss for Active Learning

2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)

IF0

2019-06-01

OAAI

Donggeun Yoo; In So Kweon

A Review on Multi-Label Learning Algorithms 多标签学习算法综述

IEEE TRANSACTIONS ON KNOWLEDGE AND DATA ENGINEERING

IF10.4

2014-08-01

2.5K

PREAI

Zhang, Min-Ling; Zhou, Zhi-Hua

NUS-WIDE

Proceedings of the ACM International Conference on Image and Video Retrieval

IF0

2009-07-08

PREAI

Tat-Seng Chua; Jinhui Tang; Richang Hong; Haojie Li; Zhiping Luo; Yantao Zheng

The Emerging Trends of Multi-Label Learning 多标签学习的新兴趋势

IEEE Transactions on Pattern Analysis and Machine Intelligence

IF0

2022-11-01

OAAI

Weiwei Liu; Haobo Wang; Xiaobo Shen; Ivor W. Tsang

Scene-Aware Label Graph Learning for Multi-Label Image Classification

2023 IEEE/CVF International Conference on Computer Vision (ICCV)

IF0

2023-10-01

PREAI

Xuelin Zhu; Jian Liu; Weijia Liu; Jiawei Ge; Bo Liu; Jiuxin Cao

Recurrent Attentional Reinforcement Learning for Multi-Label Image Recognition

Proceedings of the AAAI Conference on Artificial Intelligence

IF0

2018-04-27

OAAI

Tianshui Chen; Zhouxia Wang; Guanbin Li; Liang Lin

Explainable Reinforcement Learning: A Survey and Comparative Review

ACM COMPUTING SURVEYS

IF28

2024-04-09

OAAI

Milani, Stephanie; Topin, Nicholay; Veloso, Manuela; Fang, Fei

Lesion Location Attention Guided Network for Multi-Label Thoracic Disease Classification in Chest X-Rays

IEEE JOURNAL OF BIOMEDICAL AND HEALTH INFORMATICS

IF6.8

2020-07-01

PREAI

Chen, Bingzhi; Li, Jinxing; Lu, Guangming; Zhang, David

Human Attribute Recognition by Deep Hierarchical Contexts

Lecture Notes in Computer Science

IF0

2016-09-17

PREAI

Yining Li; Chen Huang; Chen Change Loy; Xiaoou Tang

Human-level control through deep reinforcement learning 通过深度强化学习实现人类层面的控制

Nature

IF0

2015-02-25

PREAI

Volodymyr Mnih; Koray Kavukcuoglu; David Silver; Andrei A. Rusu; Joel Veness; Marc G. Bellemare; Alex Graves; Martin Riedmiller; Andreas K. Fidjeland; Georg Ostrovski; Stig Petersen; Charles Beattie; Amir Sadik; Ioannis Antonoglou; Helen King; Dharshan Kumaran; Daan Wierstra; Shane Legg; Demis Hassabis

查看更多内容