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.

best subset + GBT

Submitted by IDEAL, Intel

best subset + GBT

Dataset Balanced Error Area Under Curve  
Train Valid Test Train Valid Test
ada 0.1721 0.198 0.1759 0.8272 0.8089 0.8237 agnostic
gina 0 0 0.0534 1 1 0.9461 agnostic
hiva 0.1765 0.11 0.2899 0.8323 0.9021 0.7123 agnostic
nova 0.0308 0.028 0.0713 0.9708 0.9587 0.9277 agnostic
sylva 0.0098 0.0073 0.0138 0.9902 0.9921 0.9862 agnostic
Overall 0.0778 0.0687 0.1209 0.9241 0.9324 0.8792 agnostic

This entry is a complete agnostic learning entry.