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.

vn2

Submitted by Vladimir Nikulin

vn2

Dataset Balanced Error Area Under Curve  
Train Valid Test Train Valid Test
ada 0.1707 0.1804 0.195 0.8313 0.8043 0.7928 prior
gina 0 0 0.0226 1 1 0.9777 prior
hiva 0.1639 0.1614 0.3019 0.9164 0.8813 0.7481 agnostic
nova 0 0 0.0601 1 1 0.9289 agnostic
sylva 0.0024 0.0029 0.0084 0.9999 0.9999 0.9917 prior
Overall 0.0674 0.0689 0.1176 0.9495 0.9371 0.8878 prior

This entry is a complete prior knowledge entry.