AI Paper Atlas

Research map / Label-efficient and robust learning / Zero-/few-shot and meta-learning

Meta-Learning With Task-Adaptive Loss Function for Few-Shot Learning

Sungyong Baik, Janghoon Choi, Heewon Kim, Dohee Cho, Jaesik Min, Kyoung Mu Lee

ICCV 2021 · Label-efficient and robust learning › Zero-/few-shot and meta-learning · approach: Meta-learning analysis and benchmarks · 168 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 cross entropy with a meta-learned loss in the inner loop (ablation)MAML 4-CONV · miniImageNet 5-way 5-shot ablationaccuracy (%): 64.99 -> 67.42Improved
add cross entropy alongside the learned loss (ablation)MAML with learned loss · miniImageNet 5-way 5-shot ablationaccuracy (%): 67.42 -> 67.86No clear effect
make the learned loss task-adaptive via affine transformation (ablation)MAML with learned loss · miniImageNet 5-way 5-shot ablationaccuracy (%): 67.42 -> 68.56Improved
add semi-supervised inner loop using unlabeled query examples (ablation)MAML with fixed learned loss · miniImageNet 5-way 5-shot ablationaccuracy (%): 67.42 -> 68.24Improved
add both task-adaptivity and semi-supervised inner loop (full MeTAL) (ablation)MAML with fixed learned loss · miniImageNet 5-way 5-shot ablationaccuracy (%): 67.42 -> 70.52Improved
add base learner weights to the task state (ablation)MeTAL with CE-only task state · miniImageNet 5-way 5-shot ablationaccuracy (%): 67.86 -> 68.66Improved
add base learner predictions to the task state (ablation)MeTAL with CE-only task state · miniImageNet 5-way 5-shot ablationaccuracy (%): 67.86 -> 67.94No clear effect
add both weights and predictions to the task state (ablation)MeTAL with CE-only task state · miniImageNet 5-way 5-shot ablationaccuracy (%): 67.86 -> 70.52Improved
apply MeTAL to MAML (4-CONV)MAML · miniImageNet 5-way 5-shotaccuracy (%): 64.99 -> 70.52Improved
apply MeTAL to MAML++ (4-CONV)MAML++ · miniImageNet 5-way 5-shotaccuracy (%): 68.32 -> 72.89Improved
apply MeTAL to ALFA + MAML (4-CONV)ALFA + MAML · miniImageNet 5-way 5-shotaccuracy (%): 69.12 -> 74.10Improved
apply MeTAL to MAML (ResNet12)MAML · miniImageNet 5-way 1-shotaccuracy (%): 58.60 -> 59.64Improved

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