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.

PSMS_SUBSAMPLE_RUN_OLD_NOVA

Submitted by H. Jair Escalante

PSMS WITH SUBSAMPLING (REDUCING THE COMPLEXITY OF PSMS), FULL MODELS WERE CONSIDERED. WE USED THE OLD NOVA CONFIGURATION.

Dataset Balanced Error Area Under Curve  
Train Valid Test Train Valid Test
ada 0.1826 0.193 0.1801 0.9083 0.8898 0.9024 agnostic
gina 0.0016 0 0.0414 1 1 0.9921 agnostic
hiva 0.0147 0.0081 0.2892 1 1 0.7592 agnostic
nova 0.0004 0.044 0.0511 1 0.997 0.9896 agnostic
sylva 0.0301 0.0307 0.0292 0.9943 0.9964 0.9949 agnostic
Overall 0.0459 0.0552 0.1182 0.9805 0.9766 0.9277 agnostic

This entry is a complete agnostic learning entry.