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.

vn1

Submitted by Vladimir Nikulin

vn1

Dataset Balanced Error Area Under Curve  
Train Valid Test Train Valid Test
ada 0.1811 0.2029 0.183 0.8309 0.8043 0.8213 agnostic
gina 0 0 0.0232 1 1 0.9762 prior
hiva 0.1735 0.1541 0.3101 0.9333 0.9311 0.741 agnostic
nova 0.0004 0 0.0483 1 1 0.962 agnostic
sylva 0.0065 0.0037 0.0078 0.9943 0.9993 0.9929 prior
Overall 0.0723 0.0721 0.1145 0.9517 0.9469 0.8987 prior

This entry is a complete prior knowledge entry.