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.

vn3

Submitted by Vladimir Nikulin

vn3

Dataset Balanced Error Area Under Curve  
Train Valid Test Train Valid Test
ada 0.1662 0.2029 0.1788 0.848 0.8006 0.8225 prior
gina 0 0 0.0237 1 1 0.9827 prior
hiva 0.1499 0.1687 0.2904 0.8842 0.8328 0.7343 agnostic
nova 0.0004 0 0.0477 1 1 0.939 agnostic
sylva 0.0006 0.0004 0.0071 1 1 0.9959 prior
Overall 0.0634 0.0744 0.1095 0.9464 0.9267 0.8949 prior

This entry is a complete prior knowledge entry.