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.

psmsx_100_10_5_2cs

Submitted by H. Jair Escalante

psmsx_100_10_5_2cs

Dataset Balanced Error Area Under Curve  
Train Valid Test Train Valid Test
ada 0.1868 0.2193 0.1963 0.8174 0.8053 0.8042 agnostic
gina 0 0 0.0352 1 1 0.965 agnostic
hiva 0.1212 0.1834 0.298 0.8816 0.7971 0.7048 agnostic
nova 0.0028 0 0.0585 0.9971 1 0.9407 agnostic
sylva 0 0 0.0726 1 1 0.9274 agnostic
Overall 0.0622 0.0805 0.1321 0.9392 0.9205 0.8684 agnostic

This entry is a complete agnostic learning entry.