Domain-Aware Graph Network for Bridging Multi-Source Domain Adaptation

计算机科学 人工智能 机器学习 特征学习 域适应 桥接(联网) 图形 学习迁移 理论计算机科学 计算机网络 分类器(UML)
作者
Jin Yuan,Feng Hou,Yang Ying,Y.S. Zhang,Zhongchao Shi,Xin Geng,Jianping Fan,Zhiqiang He,Yong Rui
出处
期刊:IEEE Transactions on Multimedia [Institute of Electrical and Electronics Engineers]
卷期号:26: 7210-7224 被引量:2
标识
DOI:10.1109/tmm.2024.3361729
摘要

Domain adaptation (DA) addresses the challenge of distribution discrepancy between the training and test data, while multi-source domain adaptation (MSDA) is particularly appealing for realistic scenarios. With the emergence of extensive unlabeled datasets, self-supervised learning has gained significant popularity in deep learning. It is noteworthy that multi-source domain adaptation and self-supervised learning share a common objective: leveraging unlabeled data to acquire more informative representations. However, conventional self-supervised learning encounters two main limitations. Firstly, the traditional pretext task falls to transfer fine-grained knowledge to downstream task with general representation learning. Secondly, the scheme of the same feature extractor with distinct prediction heads makes the cross-task knowledge exchange and information sharing ineffective. In order to tackle these challenges, we introduce a novel approach called Domain-Aware Graph Network (DAGNet). DAGNet utilizes a graph neural network as a bridge to facilitate efficient cross-task knowledge exchange. By employing a mask token strategy, we enhance the robustness of representations by selectively masking certain domain or self-supervised information. In terms of datasets, the uneven and style-based domain shifts in current datasets make it challenging to measure the model's domain adaptation performance in real-world applications. To address this issue, we introduce a benchmark dataset DomainVerse with continuous spatio-temporal domain shifts encountered in the real world. Our extensive experiments demonstrate that DAGNet achieves state-of-the-art performance not only on mainstream multi-source domain adaptation datasets but also on different settings within DomainVerse. Code is available at https://github.com/a791702141/SSG .
最长约 10秒,即可获得该文献文件

科研通智能强力驱动
Strongly Powered by AbleSci AI
科研通是完全免费的文献互助平台,具备全网最快的应助速度,最高的求助完成率。 对每一个文献求助,科研通都将尽心尽力,给求助人一个满意的交代。
实时播报
小蘑菇应助科研通管家采纳,获得10
刚刚
小马甲应助科研通管家采纳,获得10
1秒前
丘比特应助科研通管家采纳,获得10
1秒前
科研通AI5应助科研通管家采纳,获得10
1秒前
1秒前
科研通AI5应助断章采纳,获得10
1秒前
洪七公完成签到,获得积分10
1秒前
2秒前
2秒前
星辰大海应助qwer1234采纳,获得10
5秒前
学白柒发布了新的文献求助20
5秒前
5秒前
大菊发布了新的文献求助10
5秒前
张医生完成签到,获得积分10
5秒前
6秒前
冷月芳华发布了新的文献求助10
6秒前
carbonhan完成签到,获得积分10
6秒前
慕青应助Ab采纳,获得30
6秒前
干净山彤发布了新的文献求助10
7秒前
ding应助大气的天蓝采纳,获得10
9秒前
knn发布了新的文献求助10
9秒前
zzzhu发布了新的文献求助10
9秒前
9秒前
alfredwu94完成签到,获得积分10
10秒前
10秒前
10秒前
nengzou完成签到,获得积分10
10秒前
11秒前
11秒前
主旋律完成签到 ,获得积分10
12秒前
朴实的猎豹完成签到,获得积分10
12秒前
13秒前
花开富贵完成签到,获得积分10
13秒前
PG发布了新的文献求助30
14秒前
14秒前
科研通AI5应助llg采纳,获得10
14秒前
HR112发布了新的文献求助10
14秒前
14秒前
啊咧发布了新的文献求助10
14秒前
干净山彤完成签到,获得积分10
14秒前
高分求助中
Les Mantodea de Guyane Insecta, Polyneoptera 2500
Technologies supporting mass customization of apparel: A pilot project 450
China—Art—Modernity: A Critical Introduction to Chinese Visual Expression from the Beginning of the Twentieth Century to the Present Day 430
Tip60 complex regulates eggshell formation and oviposition in the white-backed planthopper, providing effective targets for pest control 400
A Field Guide to the Amphibians and Reptiles of Madagascar - Frank Glaw and Miguel Vences - 3rd Edition 400
China Gadabouts: New Frontiers of Humanitarian Nursing, 1941–51 400
The Healthy Socialist Life in Maoist China, 1949–1980 400
热门求助领域 (近24小时)
化学 材料科学 医学 生物 工程类 有机化学 物理 生物化学 纳米技术 计算机科学 化学工程 内科学 复合材料 物理化学 电极 遗传学 量子力学 基因 冶金 催化作用
热门帖子
关注 科研通微信公众号,转发送积分 3793698
求助须知:如何正确求助?哪些是违规求助? 3338599
关于积分的说明 10290546
捐赠科研通 3055010
什么是DOI,文献DOI怎么找? 1676285
邀请新用户注册赠送积分活动 804326
科研通“疑难数据库(出版商)”最低求助积分说明 761836