AI Paper Atlas

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

LegendreTron: Uprising Proper Multiclass Loss Learning

Kevin H Lam, Christian Walder, Spiridon Penev, Richard Nock

ICML 2023 · Learning theory and algorithms › Prediction and decision losses · approach: Ranking-based loss optimization · 0 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
Learn a proper multiclass canonical loss and probabilities jointly instead of using multinomial logistic regressionLegendreTron · aloi, sector, letter, news20, Sensorless, vowel, segment, vehicleclassification accuracy: aloi: 10.34$\pm$0.42 -> 88.11$\pm$0.03; sector: 8.77$\pm$0.73 -> 89.71$\pm$0.18; letter: 53.37$\pm$0.25 -> 79.82$\pm$0.3Improved
Learn a proper multiclass canonical loss and probabilities jointly instead of using multinomial logistic regressionLegendreTron · dnaclassification accuracy: 92.79$\pm$0.30 -> 94.43$\pm$0.19Hurt
Learn a proper multiclass canonical loss and probabilities jointly instead of using multinomial logistic regressionLegendreTron · svmguide2classification accuracy: 56.01$\pm$1.40 -> 56.01$\pm$1.40No clear effect
Learn a proper multiclass canonical loss and probabilities jointly instead of using multinomial logistic regressionLegendreTron · wineclassification accuracy: 96.94$\pm$1.14 -> 97.78$\pm$0.59No clear effect
Learn a proper multiclass canonical loss and probabilities jointly instead of using multinomial logistic regressionLegendreTron · irisclassification accuracy: 86.67$\pm$3.89 -> 83.00$\pm$2.08No clear effect
Learn a proper multiclass loss while retaining a VGG-5 modelVGGTron versus VGG · MNISTclassification accuracy: 99.40\% -> 99.59\%Improved
Learn a proper multiclass loss while retaining a VGG-5 modelVGGTron versus VGG · Fashion-MNISTclassification accuracy: 92.80\% -> 92.88\%Improved
Learn a proper multiclass loss while retaining a VGG-5 modelVGGTron versus VGG · Kuzushiji-MNISTclassification accuracy: 98.12\% -> 98.26\%Improved
Model all ten class probabilities jointly rather than using ten one-vs-rest binary modelsLegendreTron · MNIST-like datasetsclassification accuracyImproved
Use the learned inverse canonical link rather than the identity mapLegendreTron · multiclass classificationclassification accuracyImproved
Use softmax+ with an identity inverse-link component, removing learned loss adaptation (ablation)LegendreTron · MNISTtest AUC: 99.9\% -> 99.9\%No clear effect
Use softmax+ with an identity inverse-link component, removing learned loss adaptation (ablation)LegendreTron · Fashion-MNISTtest AUC: 99.2\% -> 98.5\%Hurt

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