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.

GBT + PF

Submitted by Vladimir Martyanov

Gradient Boosted Trees with Particle Filtering

Dataset Balanced Error Area Under Curve  
Train Valid Test Train Valid Test
ada 0.1516 0.1608 0.1752 0.8477 0.8469 0.8252 agnostic
gina 0 0 0.0415 1 1 0.9589 agnostic
hiva 0.1647 0.1025 0.2946 0.847 0.905 0.7042 agnostic
nova 0.0012 0 0.0772 0.9988 1 0.922 agnostic
sylva 0 0 0.0079 1 1 0.9921 agnostic
Overall 0.0635 0.0527 0.1193 0.9387 0.9504 0.8805 agnostic

This entry is a complete agnostic learning entry.