AI Paper Atlas

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

Contradistinguisher: A Vapnik's Imperative to Unsupervised Domain Adaptation

Sourabh Balgi, Ambedkar Dukkipati

TPAMI 2021 · Label-efficient and robust learning › Domain adaptation and generalization · approach: Conditional distribution alignment · 11 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 target unsupervised contradistinguish loss to source supervised training (ablation)CUDA · Office-31 single-source, fixed ResNet-50mean accuracy: 82.1 -> 88.7Improved
add target unsupervised loss (ablation)CUDA · Office-31 single-source, fixed ResNet-152mean accuracy: 84.6 -> 90.0Improved
add target unsupervised loss (ablation)CUDA · Office-31 multi-sourcemean accuracy: 87.6 -> 90.4Improved
add source unsupervised loss (ablation)CUDA · Office-31 single-source, fixed ResNet-50mean accuracy: 88.7 -> 88.5No clear effect
add target adversarial regularization (ablation)CUDA · Office-31 multi-sourcemean accuracy: 89.5 -> 90.8Improved
add target adversarial regularization (ablation)CUDA · Office-31 single-source, fixed ResNet-152mean accuracy: 89.7 -> 90.7Improved
add target adversarial regularization (ablation)CUDA · Office-31 single-source, fixed ResNet-50mean accuracy: 88.5 -> 88.0Hurt
add source adversarial regularization (ablation)CUDA · Office-31 single-source, fixed ResNet-50mean accuracy: 88.0 -> 88.9Improved
fix pretrained ResNet-50 instead of fine-tuning for source-only training (ablation)source supervised baseline · Office-31 single-sourcemean accuracy: 50.6 -> 82.1Improved
use ResNet-152 instead of ResNet-50 encoderCUDA · Office-31 single-sourcemean accuracy: 89.8 -> 91.1Improved
add data augmentation and target clustering like CAN (CUDA*) (ablation)CUDA · VisDA-2017 testaccuracy: 78.1 -> 85.8Improved
add data augmentation and target clustering like CAN (CUDA*) (ablation)CUDA · VisDA-2017 validationaccuracy: 79.6 -> 84.2Improved

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