AI Paper Atlas

Research map / Learning theory and algorithms / Prediction and decision losses

The Lovász Hinge: A Novel Convex Surrogate for Submodular Losses

Jiaqian Yu, Matthew B. Blaschko

TPAMI 2018 · Learning theory and algorithms › Prediction and decision losses · approach: Consistency of surrogate losses · 40 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 Lovasz hinge instead of 0-1 SVMsubmodular loss learning · synthetic early detectiontest Delta1: 0.166 -> 0.100Improved
train with Lovasz hinge instead of slack rescalingsubmodular loss learning · synthetic early detectiontest Delta1: 0.144 -> 0.100Improved
train with Lovasz hinge instead of margin rescalingsubmodular loss learning · synthetic early detectiontest Delta1: 0.154 -> 0.100Improved
train with Lovasz hinge instead of 0-1 losssubmodular loss learning · synthetic early detectiontest 0-1 loss: 2.87 -> 7.42Hurt
train with Lovasz hinge instead of 0-1 SVMsubmodular loss learning · PASCAL VOC multilabel Delta2test Delta2: 0.5091 -> 0.4371Improved
train with Lovasz hinge instead of slack rescalingsubmodular loss learning · PASCAL VOC multilabel Delta2test Delta2: 0.4927 -> 0.4371Improved
train with Lovasz hinge instead of margin rescalingsubmodular loss learning · PASCAL VOC multilabel Delta2test Delta2: 0.4437 -> 0.4371Improved
train with Lovasz hinge instead of 0-1 SVMsubmodular loss learning · PASCAL VOC multilabel Delta3test Delta3: 0.9877 -> 0.9447Improved
train with Lovasz hinge instead of margin rescalingsubmodular loss learning · PASCAL VOC multilabel Delta3test Delta3: 0.9718 -> 0.9447Improved
train with Lovasz hinge instead of 0-1 SVMsubmodular loss learning · PASCAL VOC multilabel Delta2test 0-1 loss: 0.8320 -> 0.9329Hurt
train with Lovasz hinge instead of margin rescalingsubmodular loss learning · PASCAL VOC multilabel Delta2test 0-1 loss: 1.0010 -> 0.9329Improved

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