AI Paper Atlas

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

Adversarial Bayesian Augmentation for Single-Source Domain Generalization

Sheng Cheng, Tejas Gokhale, Yezhou Yang

ICCV 2023 · Label-efficient and robust learning › Domain adaptation and generalization · approach: Foundation-model domain generalization · 33 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
replace ALT adversarial standard conv with adversarial Bayesian conv (1-layer)ABA · Digits target avgaccuracy (%): 70.23 -> 74.57Improved
replace ALT with ABA at 5 layers on PACSABA · PACS avgaccuracy (%): 63.60 -> 66.02Improved
add RandConv module to 1-layer ABA (ablation)ABA 1-layer · Digits target avgaccuracy (%): 74.57 -> 74.51No clear effect
add RandConv or Augmix module to ABA (ablation)ABA 5-layer · PACS avgaccuracy (%): 66.02 -> 66.00 (RandConv), 65.55 (Augmix)No clear effect
add Augmix module to 5-layer ABA (ablation)ABA 5-layer · Camelyon17 target avgaccuracy (%): 88.19 -> 89.88Improved
increase ABA from 1 layer to 3 layers (ablation)ABA · Digits target avgaccuracy (%): 74.57 -> 76.56Improved
increase ABA from 3 layers to 5 layers (ablation)ABA · Digits target avgaccuracy (%): 76.56 -> 74.76Hurt
adversarial learning rate 5e-6 versus 5e-5 or 5e-7 (ablation)ABA · DigitsaccuracyHurt
increase adversarial steps beyond 5 (ablation)ABA · DigitsaccuracyNo clear effect
increase number of Bayesian samplings beyond 10 (ablation)1-layer ABA · DigitsaccuracyHurt
apply AdvBNN instead of ERMbaseline · Digits target avgaccuracy (%): 53.50 -> 51.92Hurt
apply RandConv instead of ERM on subpopulation shiftbaseline · Living17 targetaccuracy (%): 70.97 -> 61.18Hurt

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