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.

Nova_1_02_300_NCV

Submitted by H. Jair Escalante

Boosting NN, 1 unit, shrinkage 0.1, 300 its

Dataset Balanced Error Area Under Curve  
Train Valid Test Train Valid Test
ada 0.1231 0.2335 0.2078 0.9594 0.844 0.877 agnostic
gina 0 0.0477 0.0623 1 0.9852 0.9839 agnostic
hiva 0.0726 0.3135 0.2949 0.9745 0.7795 0.7675 agnostic
nova 0 0.108 0.0846 1 0.9915 0.9868 agnostic
sylva 0.0072 0.0053 0.0117 0.999 0.9991 0.9983 agnostic
Overall 0.0406 0.1416 0.1323 0.9866 0.9199 0.9227 agnostic

This entry is a Model Selection Game entry.