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.

cross-indexing-8

Submitted by Juha Reunanen

Like cross-indexing-7 (and -7a), but model selection done using validation data as well.
Still using CLOP models only.

Dataset Balanced Error Area Under Curve  
Train Valid Test Train Valid Test
ada 0.1395 0.1659 0.1807 0.9444 0.9267 0.907 agnostic
gina 0 0 0.0474 1 1 0.9904 agnostic
hiva 0.0639 0.0763 0.2977 0.9792 0.9793 0.7663 agnostic
nova 0.0004 0 0.0499 1 1 0.9907 agnostic
sylva 0.004 0.0033 0.0111 0.9994 0.9996 0.9986 agnostic
Overall 0.0416 0.0491 0.1173 0.9846 0.9811 0.9306 agnostic

This entry is a complete agnostic learning entry.