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.

trees

Submitted by Joerg Wichard

decisoin trees
validation data invoked

Dataset Balanced Error Area Under Curve  
Train Valid Test Train Valid Test
ada 0.1848 0.2053 0.2437 0.9529 0.9413 0.9028 agnostic
gina 0.0604 0.0443 0.0914 0.9857 0.9932 0.9664 agnostic
hiva 0.1638 0.1512 0.2976 0.9207 0.907 0.7639 agnostic
nova 0.0066 0 0.1104 0.9998 1 0.9603 agnostic
sylva 0.0169 0.0197 0.0266 0.9997 0.9997 0.9981 agnostic
Overall 0.0865 0.0841 0.1539 0.9718 0.9682 0.9183 agnostic

This entry is a complete agnostic learning entry.