AI Paper Atlas

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

Reducing the Covariate Shift by Mirror Samples in Cross Domain Alignment

Yin Zhao, Minquan Wang, Longjun Cai

NeurIPS 2021 · Label-efficient and robust learning › Domain adaptation and generalization · approach: Statistical alignment and hypothesis testing · 5 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 mirror loss at both backbone and FC feature layers (ablation)UDA model · Office-Homeaccuracy: 66.5 -> 73.4Improved
Add mirror loss at both backbone and FC feature layers (ablation)UDA model · Office-31accuracy: 85.5 -> 91.7Improved
Add mirror loss at backbone feature layer f only (ablation)UDA model · Office-Homeaccuracy: 66.5 -> 71.8Improved
Add mirror loss at FC feature layer g only (ablation)UDA model · Office-Homeaccuracy: 66.5 -> 71.7Improved
Add mirror loss at backbone feature layer f only (ablation)UDA model · Office-31accuracy: 85.5 -> 90.0Improved
Add mirror loss at FC feature layer g only (ablation)UDA model · Office-31accuracy: 85.5 -> 89.7Improved
Use k=3 instead of k=1 for mirror-neighbor count (ablation)Mirror alignment · Office-Homeaccuracy: 71.7 -> 73.4Improved
Increase mirror-neighbor count from k=3 to k=7 (ablation)Mirror alignment · Office-Homeaccuracy: 73.4 -> 71.3Hurt
Increase mirror-neighbor count from k=3 to k=9 (ablation)Mirror alignment · Office-Homeaccuracy: 73.4 -> 71.4Hurt
Use k=3 instead of k=1 for mirror-neighbor count (ablation)Mirror alignment · Office-31accuracy: 90.2 -> 91.7Improved
Increase mirror-neighbor count from k=3 to k=7 (ablation)Mirror alignment · Office-31accuracy: 91.7 -> 89.8Hurt
Vary distance metric and mirror-sample weighting from Euclidean with 1/k weights (ablation)Mirror alignment · Office-Homeaccuracy: 72.7 to 72.3No clear effect

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