AI Paper Atlas

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

Feature-Level Domain Adaptation

Wouter M. Kouw, Laurens J.P. van der Maaten, Jesse H. Krijthe, Marco Loog

JMLR 2016 · Label-efficient and robust learning › Domain adaptation and generalization · approach: Source-free domain adaptation · 52 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
perturb estimated transfer parameter theta_1 by +0.1, +0.2, +0.3 (ablation)FLDA-Q · Synthetic Poissonclassification error: .138 -> .145 -> .149 -> .150Hurt
perturb estimated transfer parameter theta_1 by +0.1, +0.2, +0.3 (ablation)FLDA-L · Synthetic Poissonclassification error: .139 -> .140 -> .142 -> .146Hurt
adapt with dropout transfer model (FLDA-Q) instead of naive source least squaresS-LS · Synthetic Bernoulliclassification error: 0.400 -> 0.300Improved
adapt with dropout transfer model (FLDA-Q) instead of naive source least squaresS-LS · Synthetic Poissonclassification error: 0.181 -> 0.099Improved
adapt with FLDA-L instead of naive source logistic regressionS-LR · Synthetic Poissonclassification error: 0.170 -> 0.084Improved
adapt with FLDA-L instead of naive source logistic regressionS-LR · Synthetic Bernoulliclassification error: 0.248 -> 0.301Hurt
use FLDA-Q with few training samples versus naive classifiersS-LS and T-LS · Synthetic Poisson learning curvesclassification errorImproved
apply FLDA (L and Q) to missing-data-at-test-timeS-LR · UCI Heart Diseaseclassification error: .338 -> .203Improved
apply FLDA-L to missing-data-at-test-timeS-LR · UCI Hepatitisclassification error: .493 -> .200Improved
apply FLDA-LS-LR · MNIST/USPS digits M->Uclassification error: .747 -> .678No clear effect
apply FLDA-QS-LR · MNIST/USPS digits U->Mclassification error: .770 -> .640Improved
use FLDA-L instead of FLDA-QFLDA · Amazon sentiment (B->D)classification error: .303 -> .166Improved

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