AI Paper Atlas

Research map / Detection and segmentation / CNN features

SAAL: Sharpness-Aware Active Learning

Yoon-Yeong Kim, Youngjae Cho, JoonHo Jang, Byeonghu Na, Yeongmin Kim, Kyungwoo Song, Wanmo Kang, Il-chul Moon

ICML 2023 · Detection and segmentation › CNN features · approach: Local features and context in CNNs

Read the paper

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
Use SAAL acquisition (perturbed pseudo-label loss) instead of baselines with Adam optimizeractive learning on image classification · CIFAR-10test accuracy (%): Random 50.7 -> SAAL 54.4 (Adam, with k-means++ variant)Improved
Add SAM optimizer to SAAL acquisition (ablation)SAAL · CIFAR-10test accuracy (%): SAAL Adam 52.3 -> SAAL SAM 56.0; on Fashion SAM lowers SAAL 85.6 -> 85.0Mixed
Add k-means++ seeding diversity to SAAL acquisition (ablation)SAAL · CIFAR-10test accuracy (%): Adam 52.3 -> 54.4; SAM 56.0 -> 57.0 (reported as w/ k-means++)Improved
Use SAAL acquisition instead of BADGE with Adam optimizeractive learning on image classification · Fashion-MNISTtest accuracy (%): BADGE 85.4 -> SAAL 85.6Improved
Use SAAL acquisition instead of BADGE with SAM optimizeractive learning on image classification · Fashion-MNISTtest accuracy (%): BADGE 86.2 -> SAAL 85.0Hurt
Use SAAL acquisition instead of baselines on long-tailed CIFAR-10 (ablation)active learning on imbalanced data · Long-tailed CIFAR-10test accuracy (%): BADGE 21.69 -> SAAL 23.03Improved
Use SAAL score (perturbed loss) for object detection image selection with SSDSSD object detection · PASCAL VOC 2007+2012mAP: BADGE 0.7493 -> SAAL 0.7541Improved
Combine SAAL score with diversity score in region-based annotation (RI-SAAL) for domain adaptive segmentationSYNTHIA to CityScapes segmentation · SYNTHIA to CityScapesmIOU: RIPU 70.2 -> RI-SAAL 70.6Improved
Use SAM-perturbation sensitivity radius rho=0.05 vs 0.01 or 0.10 (ablation)SAAL · CIFAR-10test accuracy (%)No clear effect
Batch-wise perturbation (batch size 10) instead of per-instance perturbation in SAAL (ablation)SAAL · CIFAR-10test accuracy (%): Adam: BS1 54.4 vs BS10 54.0; SAM: 57.0 vs 57.7; wall time drops from 49.6 s to 16.8 sNo clear effect
Use SAM optimizer instead of Adam for SAAL (ablation)SAAL · CIFAR-100test accuracy (%): 46.6 -> 48.4Improved
Use SAAL acquisition in high-budget setting (1,000 per step) (ablation)active learning on image classification · Fashion, SVHN, CIFAR-10, CIFAR-100 (high budget)test accuracy: small margin over baselinesImproved

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