AI Paper Atlas

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

Gradient Harmonization in Unsupervised Domain Adaptation

Fuxiang Huang, Suqi Song, Lei Zhang

TPAMI 2024 · Label-efficient and robust learning › Domain adaptation and generalization · approach: Feature alignment for unsupervised adaptation · 28 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 GH to CDANCDAN · Office-31 Avg excl. W<->Daccuracy (%): 81.8 -> 83.3Improved
add GH++ to CDANCDAN · Office-31 Avg excl. W<->Daccuracy (%): 81.8 -> 84.0Improved
add GH to MCDMCD · Office-Home Avgaccuracy (%): 64.1 -> 67.5Improved
add GH++ to MCDMCD · Digits Avgaccuracy (%): 94.8 -> 97.2Improved
add GH++ to DWLDWL · Office-Home Avgaccuracy (%): 63.6 -> 64.6Improved
add GH to DWLDWL · Office-31 Avg excl. W<->Daccuracy (%): 80.8 -> 81.0No clear effect
add GH++ to GVBGVB · Office-31 Avg excl. W<->Daccuracy (%): 84.2 -> 85.0No clear effect
add GH++ to SSRT (ViT IN-21K)SSRT · Office-31 Avg excl. W<->Daccuracy (%): 91.7 -> 92.2Improved
add GH++ to SSRT (ViT IN-1K)SSRT* · Office-31 Avg excl. W<->Daccuracy (%): 90.5 -> 91.4Improved
add GH++ to SSRTSSRT · Office-Home Avgaccuracy (%): 85.7 -> 86.3Improved
use GH++ instead of GHCDAN · Office-Home Avgaccuracy (%): 68.3 -> 68.9Improved
use GH++ instead of GHMCD · Office-31 Avg excl. W<->Daccuracy (%): 81.0 -> 81.4Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

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