AI Paper Atlas

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

Distributional Robustness Loss for Long-Tail Learning

Dvir Samuel, Gal Chechik

ICCV 2021 · Label-efficient and robust learning › Recognition and long-tail · approach: Classifier rebalancing and margin losses · 106 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 a robustness margin with a shared radius (eps>0) (ablation)DRO-LT · CIFAR-100-LT (imbalance 100)top-1 accuracy: 43.92 -> 45.66Improved
scale the radius by 1/sqrt(class size) instead of sharing it (ablation)DRO-LT · CIFAR-100-LT (imbalance 100)top-1 accuracy: 45.66 -> 46.92Improved
learn a per-class radius instead of eps/sqrt(n) (ablation)DRO-LT · CIFAR-100-LT (imbalance 100)top-1 accuracy: 46.92 -> 47.31Improved
full learned radius vs no margin (ablation)DRO-LT · CIFAR-100-LT (imbalance 50)top-1 accuracy: 52.31 -> 57.57Improved
full learned radius vs no margin (ablation)DRO-LT · CIFAR-100-LT (imbalance 10)top-1 accuracy: 59.54 -> 63.41Improved
full learned radius vs no margin (ablation)DRO-LT · ImageNet-LTtop-1 accuracy: 48.1 -> 53.5Improved
replace CE training with DRO-LT (learned eps)ResNet-32 · CIFAR-100-LT (imbalance 100), Few classesfew-shot accuracy: 7.4 -> 23.8Improved
replace CE training with DRO-LT (learned eps)ResNet-32 · CIFAR-100-LT (imbalance 100), Many classesmany-shot accuracy: 65.5 -> 64.7No clear effect
use DRO-LT representation instead of CE representation, with cRT classifier (ablation)ResNet-32 · CIFAR-100-LT (imbalance 100)top-1 accuracy: 41.2 -> 47.3Improved
use DRO-LT representation instead of CE representation, with CE classifier (ablation)ResNet-32 · CIFAR-100-LTtop-1 accuracy: 38.5 -> 41.2Improved
use DRO-LT representation instead of CE representation, with LWS classifier (ablation)ResNet-32 · CIFAR-100-LTtop-1 accuracy: 41.4 -> 46.8Improved
learn representation with balanced resampling instead of CE (ablation)ResNet-32 · CIFAR-100-LT (CE classifier)top-1 accuracy: 38.5 -> 34.9Hurt

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