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.

TMK B1

Submitted by The Machine

TMK B1
short run

Dataset Balanced Error Area Under Curve  
Train Valid Test Train Valid Test
ada 0.1745 0.1948 0.1802 0.9106 0.8804 0.9038 agnostic
gina 0.0749 0.0884 0.098 0.9804 0.9752 0.9618 agnostic
hiva 0.165 0.3305 0.3165 0.9282 0.7386 0.7496 agnostic
nova 0.1303 0.144 0.1509 0.9199 0.8928 0.9091 agnostic
sylva 0.0084 0.0176 0.0135 0.9992 0.9973 0.998 agnostic
Overall 0.1106 0.1551 0.1518 0.9477 0.8969 0.9044 agnostic

This entry is a complete agnostic learning entry.