AI Paper Atlas

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

Reducing Network Agnostophobia

Akshay Raj Dhamija, Manuel Günther, Terrance Boult

NeurIPS 2018 · Label-efficient and robust learning › Recognition and long-tail · approach: Group-robust classification · 79 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
Train with Objectosphere loss instead of a background classLeNet++ · DevanagariCCR at FPR 10^-2: 0.7527 -> 0.9658Improved
Train with Objectosphere loss instead of a background classLeNet++ · NotMNISTCCR at FPR 10^-2: 0.9068 -> 0.9563Improved
Train with Entropic Open-Set loss instead of a background classLeNet++ · CIFAR-10 unknownsCCR at FPR 10^-2: 0.9726 -> 0.9804Improved
Use Scaled Objectosphere instead of a background classResNet-18 · SVHNCCR at FPR 10^-2: 0.4803 -> 0.5454Improved
Use Scaled Objectosphere instead of a background classResNet-18 · CIFAR-100 subsetCCR at FPR 10^-2: 0.3429 -> 0.4334Improved
Add a feature-magnitude margin for known samples and penalize unknown feature magnitude (ablation)Entropic Open-Set on LeNet++ · CIFAR-10 unknownsCCR at FPR 10^-2: 0.9804 -> 0.9785Hurt
Add the Objectosphere feature-magnitude constraint (ablation)Entropic Open-Set on LeNet++ · NotMNISTCCR at FPR 10^-2: 0.9515 -> 0.9563Improved
Scale the softmax score by deep feature magnitude (ablation)Objectosphere on ResNet-18 · SVHNCCR at FPR 10^-2: 0.5074 -> 0.5454Improved
Scale the softmax score by deep feature magnitude (ablation)Objectosphere on ResNet-18 · CIFAR-100 subsetCCR at FPR 10^-2: 0.3595 -> 0.4334Improved
Scale the softmax score by deep feature magnitude (ablation)Objectosphere on LeNet++ · CIFAR-10 unknownsCCR at FPR 10^-2No clear effect
Train with CIFAR samples as negative unknowns instead of NIST Letters (ablation)LeNet++ · NIST Letters unknownsrobustness to unknownsHurt

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