AI Paper Atlas

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

Maximum Density Divergence for Domain Adaptation

Jingjing Li, Erpeng Chen, Zhengming Ding, Lei Zhu, Ke Lü, Heng Tao Shen

TPAMI 2020 · Label-efficient and robust learning › Domain adaptation and generalization · approach: Adversarial domain adaptation · 346 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 MDD loss to adversarial network (ablation)CDAN · SVHN->MNISTaccuracy: 89.2 -> 96.1Improved
add MDD loss (ablation)CDAN · MNIST->SVHNaccuracy: 71.3 -> 76.6Improved
add MDD loss (ablation)CDAN · MNIST->USPSaccuracy: 95.6 -> 96.1Improved
add MDD loss (ablation)CDAN · Office-31 (ResNet-50) average 2accuracy: 81.8 -> 84.9Improved
add MDD loss (ablation)CDAN · Office-31 (AlexNet) average 1accuracy: 77.7 -> 79.6Improved
add MDD loss (ablation)CDAN · ImageCLEF-DAaccuracy: 87.7 -> 90.0Improved
add MDD loss (ablation)CDAN · Office-Homeaccuracy: 65.8 -> 67.9Improved
add MDD loss (ablation)CDAN · Office-31 W->Daccuracy: 100 -> 100No clear effect
add all three intra-MDD terms instead of none (ablation)ATM · SVHN->MNISTaccuracy: 89.2 -> 96.1Improved
add term 3 to terms 1 and 2 (ablation)ATM · SVHN->MNISTaccuracy: 95.3 -> 96.1Improved
add term 2 to terms 1 and 3 (ablation)ATM · SVHN->MNISTaccuracy: 95.0 -> 96.1Improved
add term 1 to terms 2 and 3 (ablation)ATM · SVHN->MNISTaccuracy: 93.8 -> 96.1Improved

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