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.

LogitBoost with trees

Submitted by Roman Lutz

LogitBoost with trees

Dataset Balanced Error Area Under Curve  
Train Valid Test Train Valid Test
ada 0.1573 0.1691 0.1685 0.842 0.8471 0.83 agnostic
gina 0 0 0.0339 1 1 0.9668 agnostic
hiva 0.0465 0.0419 0.3038 0.953 0.9579 0.7019 agnostic
nova 0 0 0.0463 1 1 0.9538 agnostic
sylva 0.0036 0.0049 0.0062 0.9965 0.9942 0.9938 agnostic
Overall 0.0415 0.0432 0.1117 0.9583 0.9598 0.8892 agnostic

This entry is a complete agnostic learning entry.