AI Paper Atlas

Research map / Learning theory and algorithms / Kernels and regression theory

A Refined Margin Distribution Analysis for Forest Representation Learning

Shen-Huan Lyu, Liang Yang, Zhi-Hua Zhou

NeurIPS 2019 · Learning theory and algorithms › Kernels and regression theory · approach: Statistical perspectives on generalization · 7 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
remove PRECONC (pass only input features to next layer) (ablation)mdDF · PROTEINtest accuracy: 71.247 -> 68.509Hurt
remove PRECONC (pass only input features to next layer) (ablation)mdDF · SENSITtest accuracy: 82.534 -> 80.320Hurt
remove PRECONC (pass only input features to next layer) (ablation)mdDF · ADULTtest accuracy: 86.560 -> 85.650Hurt
stacking: pass only prediction vectors to next layer (ablation)mdDF · SENSITtest accuracy: 82.534 -> 80.412Hurt
stacking: pass only prediction vectors to next layer (ablation)mdDF · PROTEINtest accuracy: 71.247 -> 70.291Hurt
use same forests (4 random forests) in each layer (ablation)mdDF · ADULTtest accuracy: 86.560 -> 86.200Hurt
use same forests (4 random forests) in each layer (ablation)mdDF · LETTERtest accuracy: 97.500 -> 96.475Hurt
use same forests (4 random forests) in each layer (ablation)mdDF · MNISTtest accuracy: 98.734 -> 98.254Hurt
margin distribution reweighting and boosted representations vs gcForestgcForest · ADULTtest accuracy: 86.276 -> 86.560Improved
margin distribution reweighting and boosted representations vs gcForestgcForest · MNISTtest accuracy: 98.252 -> 98.734Improved
margin distribution reweighting vs gcForestgcForest · PROTEINtest accuracy: 71.009 -> 71.247Improved
deep forest (mdDF) instead of MLPMLP · ADULTtest accuracy: 80.597 -> 86.560Improved

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