AI Paper Atlas

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

Generalized Boosting

Arun Suggala, Bingbin Liu, Pradeep K. Ravikumar

NeurIPS 2020 · Learning theory and algorithms › Prediction and decision losses · approach: Data attribution and recourse · 0 citations (OpenAlex)

Read the paper

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
Use dense connections to earlier layers (DenseCompBoost vs StdCompBoost) (ablation)compositional boosting · Simulation 3test accuracy: 82.49 -> 95.70Improved
Use dense connections (DenseCompBoost vs StdCompBoost) (ablation)compositional boosting · Simulation 1test accuracy: 91.53 -> 93.55Improved
Increase complexity of weak transformer with depth (CmplxCompBoost vs StdCompBoost) (ablation)compositional boosting · Simulation 1test accuracy: 91.53 -> 91.97Improved
Increase complexity of weak transformer with depth (CmplxCompBoost vs StdCompBoost) (ablation)compositional boosting · CIFAR10test accuracy: 81.93 -> 82.43No clear effect
Compositional boosting instead of additive feature boostingStdCompBoost · SVHNtest accuracy: 83.36 -> 90.81Improved
Compositional boosting instead of additive feature boostingStdCompBoost · Simulation 3test accuracy: 73.13 -> 82.49Improved
Compositional boosting instead of XGBoost treesDenseCompBoost · CIFAR10test accuracy: 58.34 -> 82.31Improved
DenseCompBoost instead of XGBoost on Simulation 3DenseCompBoost · Simulation 3test accuracy: 50.10 -> 95.70Improved
Use DenseCompBoost instead of XGBoost on LetterDenseCompBoost · Lettertest accuracy: 96.16 -> 96.96No clear effect
Use DenseCompBoost instead of XGBoost on CovTypeDenseCompBoost · CovTypetest accuracy: 97.46 -> 96.28Hurt
Use CmplxCompBoost instead of End-to-End trainingCmplxCompBoost · CIFAR10test accuracy: 86.88 -> 82.43Hurt
Use DenseCompBoost instead of End-to-End trainingDenseCompBoost · FashionMNISTtest accuracy: 93.49 -> 93.17No clear effect

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