AI Paper Atlas

Research map / Label-efficient and robust learning / Domain adaptation and generalization

Generalized Domain Conditioned Adaptation Network

Shuang Li, Binhui Xie, Qiuxia Lin, Chi Harold Liu, Gao Ming Huang, Guoren Wang

TPAMI 2021 · Label-efficient and robust learning › Domain adaptation and generalization · approach: Adversarial domain adaptation · 71 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add GDCAN adaptive attention module (vs w/o AAM) (ablation)GDCAN · Office-Homeaverage accuracy: 69.2 -> 71.5Improved
add first feature adaptation module with regularization (ablation)GDCAN · Office-Homeaverage accuracy: 66.7 -> 71.5Improved
add second feature adaptation module with regularization (ablation)GDCAN · Office-Homeaverage accuracy: 68.3 -> 71.5Improved
add feature alignment modules (all) (ablation)GDCAN · Office-Homeaverage accuracy: 59.0 -> 71.5Improved
add regularization loss at first module (ablation)GDCAN · Office-Homeaverage accuracy: 68.8 -> 71.5Improved
add regularization loss at second module (ablation)GDCAN · Office-Homeaverage accuracy: 70.4 -> 71.5Improved
add target entropy loss (ablation)GDCAN · Office-Homeaverage accuracy: 70.7 -> 71.5Improved
make channel attention adaptive (GDCAN over DCAN)GDCAN · DomainNet ResNet-101average accuracy: 32.4 -> 34.7Improved
make channel attention adaptive (GDCAN over DCAN)GDCAN · DomainNet ResNet-50average accuracy: 29.2 -> 32.2Improved
make channel attention adaptive (GDCAN over DCAN)GDCAN · Office-31average accuracy: 89.5 -> 89.7No clear effect
make channel attention adaptive (GDCAN over DCAN)GDCAN · ImageCLEF-DAaverage accuracy: 88.3 -> 88.7No clear effect
plug DCAN / GDCAN into CDANCDAN · Office-Homeaverage accuracy: 63.8 -> 69.6 -> 70.3Improved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy