AI Paper Atlas

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

Boosting the Generalization Capability in Cross-Domain Few-Shot Learning via Noise-Enhanced Supervised Autoencoder

Hanwen Liang, Qiong Zhang, Peng Dai, Juwei Lu

ICCV 2021 · Label-efficient and robust learning › Domain adaptation and generalization · approach: Cross-domain few-shot with transformers · 78 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 classification loss on reconstructed images (SAE -> NSAE), CE+CE (ablation)NSAE ResNet10 · ISIC 5-shotaccuracy: 52.28 -> 54.05Improved
add classification loss on reconstructed images (SAE -> NSAE), CE+CE (ablation)NSAE Conv4 · CUB 5-shotaccuracy: 52.74 -> 55.35Improved
add classification loss on reconstructed images instead of doubling original-image loss (SAE(*) -> NSAE) (ablation)NSAE ResNet10 · ISIC 5-shotaccuracy: 52.11 -> 54.05Improved
use two-step fine-tuning instead of one-step (NSAE(-) -> NSAE), CE+CE (ablation)NSAE ResNet10 · ISIC 5-shotaccuracy: 53.52 -> 54.05Improved
pre-train with NSAE instead of a single encoder, CE+CEFine-tune baseline · ISIC 5-shotaccuracy: 48.11 -> 54.05Improved
pre-train with NSAE instead of a single encoder, BSR+CEBSR baseline · EuroSAT 5-shotaccuracy: 80.89 -> 84.33Improved
pre-train with NSAE instead of single encoder, CE+CEFine-tune baseline · CUB 5-shotaccuracy: 64.14 -> 68.51Improved
pre-train with NSAE instead of single encoder, CE+CEFine-tune baseline · CUB 50-shotaccuracy: 89.61 -> 89.42No clear effect
add data augmentation and label propagation to NSAE (BSR+CE)NSAE (BSR+CE) · ISIC 5-shotaccuracy: 55.27 -> 56.85Improved
add data augmentation and label propagation to NSAE (BSR+CE)NSAE (BSR+CE) · ChestX 50-shotaccuracy: 38.52 -> 41.80Mixed
use distance-based fine-tuning instead of CE fine-tuning when pre-training with CE (ablation)CE pre-training ResNet10 · ISIC 5-shotaccuracy: 51.28 -> 51.62Mixed
use distance-based fine-tuning instead of CE fine-tuning when pre-training with BSR (ablation)BSR pre-training ResNet10 · EuroSAT 5-shotaccuracy: 80.89 -> 80.13Hurt

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