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.

vrs5

Submitted by Vladimir Nikulin

version5

Dataset Balanced Error Area Under Curve  
Train Valid Test Train Valid Test
ada 0.1835 0.1933 0.1852 0.8132 0.8021 0.8032 agnostic
gina 0 0 0.0563 0.3449 0.3142 0.4159 agnostic
hiva 0.2711 0.2685 0.326 0.7197 0.7542 0.6862 agnostic
nova 0 0 0.0771 1 1 0.9077 agnostic
sylva 0.013 0.0156 0.0147 0.9794 0.9742 0.9799 agnostic
Overall 0.0935 0.0955 0.1319 0.7715 0.7689 0.7586 agnostic

This entry is a complete agnostic learning entry.