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.

IDEAL

Submitted by IDEAL, Intel

ACE FS+redundancy elimination + GBT

Dataset Balanced Error Area Under Curve  
Train Valid Test Train Valid Test
ada 0.1756 0.2029 0.1772 0.8216 0.8079 0.8225 agnostic
gina 0 0.0351 0.0516 1 0.9625 0.9486 agnostic
hiva 0.2059 0.3008 0.307 0.8033 0.7083 0.6946 agnostic
nova 0.037 0.112 0.0809 0.9647 0.8715 0.918 agnostic
sylva 0.0093 0.0197 0.0159 0.9908 0.9797 0.9841 agnostic
Overall 0.0856 0.1341 0.1265 0.9161 0.866 0.8736 agnostic

This entry is a complete agnostic learning entry.