AI Paper Atlas

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

Weighted Oblique Decision Trees

Bin-Bin Yang, Song-Qing Shen, Wei Gao

AAAI 2019 · Learning theory and algorithms › Prediction and decision losses · approach: Optimal decision tree learning · 32 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
use random initialization with weighted soft-split entropy objective (WODT) instead of random-init CO2roblique decision tree · 20 UCI/LIBSVM datasetstest accuracy: win/tie/loss 15/5/0Improved
WODT vs OC1r (random init)oblique decision tree · 20 UCI/LIBSVM datasetstest accuracy: win/tie/loss 11/7/2Improved
WODT with random init vs axis-parallel-initialized CO2oblique decision tree · 20 UCI/LIBSVM datasetstest accuracy: win/tie/loss 20/0/0Improved
WODT vs axis-parallel tree APDTdecision tree · 20 UCI/LIBSVM datasetstest accuracy: win/tie/loss 17/2/1Mixed
WODT vs random-init oblique trees on leaf countoblique decision tree · 20 datasetsleaf count: win/tie/loss 19/0/1 vs CO2rImproved
WODT vs APDT on tree sizedecision tree · 20 datasetsleaf count: win/tie/loss 14/0/6Mixed
WODT vs other trees in training time on high-dimensional data (>200 features)decision tree · usps, protein, mnisttraining time: about 100 times fasterImproved
WODT vs APDT across tree depths 2 to 20decision tree · usps, protein, connect4, mnisttest accuracyImproved

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