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.

lastent

Submitted by H. Jair Escalante

lastent

Dataset Balanced Error Area Under Curve  
Train Valid Test Train Valid Test
ada 0.1934 0.2027 0.1863 0.8106 0.8054 0.8138 agnostic
gina 0 0 0.0256 1 1 0.9748 agnostic
hiva 0.0051 0.0014 0.3085 0.9953 0.9996 0.6977 agnostic
nova 0.0573 0.032 0.0662 0.9447 0.9685 0.9334 agnostic
sylva 0.0052 0.0033 0.0105 0.9942 0.9962 0.9895 agnostic
Overall 0.0522 0.0479 0.1194 0.949 0.9539 0.8818 agnostic

This entry is a complete agnostic learning entry.