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.

mlp+commitee+mcic

Submitted by Namgil Lee

mlp+commitee+mcic submitted on 0708020925

Dataset Balanced Error Area Under Curve  
Train Valid Test Train Valid Test
ada 0.1777 0.2046 0.198 0.8977 0.883 0.888 agnostic
gina 0.105 0.1014 0.1244 0.9636 0.9655 0.9466 agnostic
hiva 0.0349 0.0351 0.2945 0.9899 0.9805 0.7446 agnostic
nova 0.0495 0.046 0.0773 0.989 0.9966 0.9753 agnostic
sylva 0.0071 0.0053 0.0098 0.9994 0.9994 0.9987 agnostic
Overall 0.0749 0.0785 0.1408 0.9679 0.965 0.9106 agnostic

This entry is a complete agnostic learning entry.