PASCAL Agnostic Learning
vs.
Prior Knowledge
IJCNN07

The challenge is now over. But it remains open for post-challenge submissions!


IMPORTANT: Entries made since February 1st 2007 might be using validation data, now available for training.

ensemble of trees

Submitted by Joerg Wichard

cross validated trees
datasets are balanced
up to 50 CV-Folds

Dataset Balanced Error Area Under Curve  
Train Valid Test Train Valid Test
ada 0.1747 0.2128 0.1888 0.9171 0.8726 0.9039 agnostic
gina 0 0.0412 0.0613 1 0.9922 0.9847 agnostic
hiva 0.1485 0.2826 0.3058 0.9421 0.778 0.7622 agnostic
nova 0.0333 0.116 0.1067 0.995 0.9382 0.9476 agnostic
sylva 0.0039 0.0115 0.0062 0.9998 0.9991 0.9988 agnostic
Overall 0.0721 0.1328 0.1338 0.9708 0.916 0.9194 agnostic

This entry is a complete agnostic learning entry.