AI Paper Atlas

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

Differentiable Top-k Classification Learning

Felix Petersen, Hilde Kuehne, Christian Borgelt, Oliver Deussen

ICML 2022 · Label-efficient and robust learning › Recognition and long-tail · approach: Learning from unlabeled and positive-unlabeled data · 10 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 a single top-1 objective with a mixture of top-1 and top-5 objectivesResNet18 with DiffSortNets · CIFAR-100Top-1 accuracy: 61.27 -> 62.00Improved
replace a single top-1 objective with a mixture of top-1 and top-5 objectivesResNet18 with SinkhornSort · CIFAR-100Top-5 accuracy: 85.31 -> 86.94Improved
use only a top-5 loss instead of mixing top-1 and top-5 (ablation)ResNet18 with SinkhornSort · CIFAR-100Top-1 accuracy: 61.27 -> 55.62Hurt
use only a top-5 loss instead of mixing top-1 and top-5 (ablation)ResNet18 with SinkhornSort · CIFAR-100Top-5 accuracy: 85.31 -> 87.04Improved
replace softmax cross-entropy with a mixture of top-1 and top-5 lossResNeXt-101 32x48d WSL with SinkhornSort · ImageNet-1KTop-5 accuracy: 97.795 -> 97.971Improved
replace softmax cross-entropy with a mixture of top-1 and top-5 lossResNeXt-101 32x48d WSL with DiffSortNets · ImageNet-21K-PTop-1 accuracy: 39.29 -> 40.22Improved
increase top-5 weight from .5 to .75 (ablation)Top-k DiffSortNets · ImageNet-21K-PTop-1 accuracy: 40.22 -> 39.56Hurt
increase top-5 weight from .5 to .9 (ablation)Top-k DiffSortNets · ImageNet-21K-PTop-1 accuracy: 40.22 -> 38.48Hurt
use uniform weights for top-1 through top-5 rather than equal top-1/top-5 weights (ablation)Top-k DiffSortNets · ImageNet-21K-PTop-1 accuracy: 40.22 -> 39.69Hurt
use a top-20 component instead of a top-5 component (ablation)Top-k DiffSortNets with equal top-1/top-k weights · ImageNet-21K-PTop-1 accuracy: 40.22 -> 39.54Hurt
rank a much larger set of classes differentially, such as m=1,000 or more (ablation)Differentiable sorting methods · ImageNet-1KaccuracyHurt

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