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.

mixed ensembles #3

Submitted by Joerg Wichard

as #2 with ratio
test:train = 1:1

Dataset Balanced Error Area Under Curve  
Train Valid Test Train Valid Test
ada 0.1746 0.2013 0.1788 0.911 0.8804 0.9065 agnostic
gina 0.0561 0.0825 0.0925 0.9902 0.9773 0.9689 agnostic
hiva 0.11 0.2604 0.2951 0.9739 0.7776 0.7651 agnostic
nova 0.1025 0.154 0.1351 0.9536 0.939 0.9143 agnostic
sylva 0.0044 0.0107 0.0074 0.9997 0.9992 0.9989 agnostic
Overall 0.0895 0.1418 0.1418 0.9657 0.9147 0.9107 agnostic

This entry is a complete agnostic learning entry.