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.

Corrida_final

Submitted by H. Jair Escalante

PSMS, 500 iterations for ADA, 100 iterations for GINA and HIVA,
SYLVA and NOVA as before. No cross validation for training the model.

Dataset Balanced Error Area Under Curve  
Train Valid Test Train Valid Test
ada 0.1759 0.2013 0.1827 0.9061 0.8763 0.9009 agnostic
gina 0.0266 0.0064 0.0614 0.9963 0.9998 0.9804 agnostic
hiva 0.0641 0.0486 0.2854 0.9841 0.9871 0.7551 agnostic
nova 0.0004 0.044 0.0511 1 0.997 0.9896 agnostic
sylva 0.0038 0.0045 0.0122 0.9997 0.9987 0.9957 agnostic
Overall 0.0542 0.061 0.1186 0.9772 0.9718 0.9243 agnostic

This entry is a complete agnostic learning entry.