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.

boosted trees 2

Submitted by Joerg Wichard

boosted trees
like boosted trees 1
but exchanged nova

Dataset Balanced Error Area Under Curve  
Train Valid Test Train Valid Test
ada 0.1807 0.1997 0.1852 0.9024 0.8781 0.9015 agnostic
gina 0.0184 0.0064 0.0731 0.9991 1 0.979 agnostic
hiva 0.1568 0.2813 0.3003 0.9245 0.7809 0.7641 agnostic
nova 0.1189 0.128 0.1416 0.9545 0.9194 0.9344 agnostic
sylva 0.0012 0.0004 0.0195 1 1 0.9989 agnostic
Overall 0.0952 0.1231 0.144 0.9561 0.9157 0.9156 agnostic

This entry is a complete agnostic learning entry.