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.

interim all prior

Submitted by reference (gcc)

interim all prior

Dataset Balanced Error Area Under Curve  
Train Valid Test Train Valid Test
ada 0.1332 0.174 0.17 0.9479 0.891 0.9149 prior
gina 0 0.0064 0.0233 1 0.9999 0.9969 prior
hiva 0.0166 0.2467 0.271 0.9989 0.6786 0.7667 prior
nova 0.0004 0.036 0.0471 1 0.9878 0.9886 prior
sylva 0.0024 0.0041 0.0059 0.9999 0.9992 0.999 prior
Overall 0.0305 0.0934 0.1035 0.9893 0.9113 0.9332 prior

This entry is a complete prior knowledge entry.