AI Paper Atlas

Research map / Label-efficient and robust learning / Recognition and long-tail

Learning Compositional Representations for Few-Shot Recognition

Pavel Tokmakov, Yu-Xiong Wang, Martial Hebert

ICCV 2019 · Label-efficient and robust learning › Recognition and long-tail · approach: Contrastive representation theory · 125 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 compositionality regularization (soft, orthogonal) (ablation)cosine classifier · CUB novel 1-shottop-5 accuracy: 46.1 -> 52.5Improved
add compositionality regularization (soft, orthogonal) (ablation)cosine classifier · SUN novel 1-shottop-5 accuracy: 35.4 -> 43.4Improved
add compositionality regularization on all classes (ablation)cosine classifier · CUB novel+base 1-shottop-5 accuracy: 58.2 -> 62.6Improved
use cosine instead of linear classifier (ablation)compositional representation · CUB novel 1-shottop-5 accuracy: 47.0 -> 52.5Improved
add data augmentation in few-shot training (ablation)Cos w/ comp · CUB novel 1-shottop-5 accuracy: 52.5 -> 53.6Improved
add data augmentation in few-shot training (ablation)Cos baseline · SUN novel 1-shottop-5 accuracy: 35.4 -> 39.9Improved
use hard sum constraint instead of none (ablation)baseline · SUN noveltop-5 accuracyHurt
use soft constraint instead of hard constraint (ablation)compositional regularization · CUB and SUN noveltop-5 accuracyImproved
add orthogonality constraint on attribute embeddings (ablation)soft compositional regularization · CUB and SUN noveltop-5 accuracyImproved
reduce attributes to 25% on CUB and 50% on SUN (ablation)compositional regularization · CUB and SUN noveltop-5 accuracyNo clear effect
add compositional representation to Matching Networks (ablation)Matching Networks · CUB novel 1-shottop-5 accuracy: 48.5 -> 50.9Improved
add compositional representation to Prototypical Networks (ablation)Prototypical Networks · CUB novel 1-shottop-5 accuracy: 43.2 -> 41.5Hurt

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