AI Paper Atlas

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

Improving Generalization With Domain Convex Game

Fangrui Lv, Jian Liang, Shuang Li, Jinming Zhang, Di Liu

CVPR 2023 · Label-efficient and robust learning › Domain adaptation and generalization · approach: Foundation-model domain generalization · 6 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
Use DCG instead of the strongest reported baselineResNet-18 DG · PACS leave-one-domain-outaverage accuracy: 85.47 -> 86.30Improved
Use DCG instead of the strongest reported baselineResNet-50 DG · PACS leave-one-domain-outaverage accuracy: 89.11 -> 89.84Improved
Use DCG instead of FACTResNet-18 DG · Office-Home leave-one-domain-outaverage accuracy: 66.56 -> 67.05Improved
Use DCG instead of CORALResNet-18 DG · mini-DomainNet leave-one-domain-outaverage accuracy: 62.87 -> 65.18Improved
Add Fourier augmentation to original-source supervision (ablation)PACS baseline · PACS leave-one-domain-outaverage accuracy: 79.9 -> 83.4Improved
Add supermodularity regularization rather than MAML meta-loss (ablation)Fourier-augmented PACS model · PACS leave-one-domain-outaverage accuracy: 84.9 -> 85.6Improved
Add supermodularity-based sample filtering (ablation)Fourier-augmented PACS model without regularization · PACS leave-one-domain-outaverage accuracy: 83.4 -> 84.4Improved
Add supermodularity regularization and filtering together rather than use MAML loss and filtering (ablation)Fourier-augmented PACS model · PACS leave-one-domain-outaverage accuracy: 85.7 -> 86.3Improved
Randomly split augmented domains between meta-train and meta-test (ablation)DCG meta-domain splitting · PACS leave-one-domain-outaverage accuracy: 86.3 -> 85.9Hurt
Limit sample filtering to augmented samples (ablation)DCG sample filter · PACS leave-one-domain-outaverage accuracy: 86.3 -> 86.1Hurt
Limit sample filtering to original samples (ablation)DCG sample filter · PACS leave-one-domain-outaverage accuracy: 86.3 -> 86.0Hurt
Increase domain diversity through augmentation without DCGBaseline methods · PACS Cartoon and Sketch target domainsaccuracyMixed

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