AI Paper Atlas

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

Deeper, Broader and Artier Domain Generalization

Da Li, Yongxin Yang, Yi-Zhe Song, Timothy M. Hospedales

ICCV 2017 · Label-efficient and robust learning › Domain adaptation and generalization · approach: Foundation-model domain generalization · 1,352 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 2-hot tensor weight generation on final layer (ablation)Ours-Full (low-rank domain-generalization CNN) · PACSaverage accuracy (%): 62.30 -> 62.62Improved
add low-rank decomposition of final layer (ablation)Ours-Full (low-rank domain-generalization CNN) · PACSaverage accuracy (%): 62.62 -> 64.10Improved
apply the DG strategy at every layer instead of the last only (ablation)Ours-Full (low-rank domain-generalization CNN) · PACSaverage accuracy (%): 64.10 -> 69.21Improved
use low-rank DG model instead of 1HNN with same featuresOurs-Full (low-rank domain-generalization CNN) · PACSaverage accuracy (%): 64.29 -> 64.65Improved
use low-rank DG model instead of 1HNN with same featuresOurs-Full (low-rank domain-generalization CNN) · VLCSaverage accuracy (%): 65.46 -> 69.65Improved
replace Deep-All fine-tuning with Ours-FullOurs-Full (low-rank domain-generalization CNN) · PACSaverage accuracy (%): 67.05 -> 69.21Improved
replace Deep-All fine-tuning with Ours-FullOurs-Full (low-rank domain-generalization CNN) · VLCSaverage accuracy (%): 72.02 -> 72.11No clear effect
replace Deep-All fine-tuning with Ours-FullOurs-Full (low-rank domain-generalization CNN) · VLCS (unseen Sun)accuracy (%): 64.16 -> 61.32Hurt
replace Deep-All fine-tuning with Ours-FullOurs-Full (low-rank domain-generalization CNN) · PACS (unseen Art painting)accuracy (%): 63.30 -> 62.86Hurt
replace Undo-Bias with Ours-MLP on SIFT-BOW featuresOurs-Full (low-rank domain-generalization CNN) · VLCS mean AP, Caraverage precision (%): 60.28 -> 62.37Improved
replace Undo-Bias with Ours-MLP on SIFT-BOW featuresOurs-Full (low-rank domain-generalization CNN) · VLCS mean AP, Birdaverage precision (%): 15.29 -> 13.72Hurt
fine-tune a CNN on aggregated sources (Deep-All) instead of prior DG methods on DeCAF6Ours-Full (low-rank domain-generalization CNN) · VLCSaverage accuracy (%): 68.60 -> 72.02Improved

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