AI Paper Atlas

Research map / Label-efficient and robust learning / Semi-supervised learning

Meta-Learning with Less Forgetting on Large-Scale Non-stationary Task Distributions

Zhenyi Wang, Li Shen, Le Fang, Qiuling Suo, Donglin Zhan, Tiehang Duan, Mingchen Gao

ECCV 2022 · Label-efficient and robust learning › Semi-supervised learning · approach: Self-training theory

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 MI regularization (ablation)ORDER · 5-shot, 250 OODaccuracy: 46.25 -> 50.23Improved
add OT regularization (labeled features only) (ablation)ORDER · 5-shot, 250 OODaccuracy: 46.25 -> 47.69Improved
add unlabeled ID features to OT loss (ablation)ORDER (OT) · 5-shot, 250 OODaccuracy: 47.69 -> 48.57Improved
add OT on top of MI (ablation)ORDER · 5-shot, 250 OODaccuracy: 50.23 -> 51.87Improved
add OT on top of MI (ablation)ORDER · 5-shot, 50 OODaccuracy: 52.63 -> 53.96Improved
filter OOD data out of unlabeled set (ablation)ORDER baseline · 5-shot, 250 OODaccuracy: 46.25 -> 48.36Improved
add MI on ID data after filtering OOD (ablation)ORDER · 5-shot, 250 OODaccuracy: 48.36 -> 49.98Improved
keep OOD data and add MI with both terms (ablation)ORDER · 5-shot, 250 OODaccuracy: 49.98 -> 51.87Improved
use unlabeled data in meta-learning (MSKM vs ProtoNet) (ablation)ProtoNet · SETS 5-shotaccuracy: 42.62 -> 45.41Improved
use unlabeled data (LST vs ANIL) (ablation)ANIL · SETS 1-shotaccuracy: 30.58 -> 34.81Improved
add OT regularization with replay vs best CL baselineSemi-MER · SETS 1-shotaccuracy: 37.90 -> 39.21Improved
add MI on top of OT (full ORDER vs CL baselines)Semi-MER · SETS 5-shotaccuracy: 50.46 -> 53.96Improved

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