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.

SVM

Submitted by Chloe Azencott

SVM

Dataset Balanced Error Area Under Curve  
Train Valid Test Train Valid Test
ada 0.2721 0.2916 0.2751 0.8158 0.7918 0.8084 agnostic
gina 0.1938 0.1784 0.1984 0.894 0.9129 0.8915 agnostic
hiva   0 0.2693   1 0.7643 prior
nova 0.1781 0.19 0.2005 0.9657 0.9685 0.9574 agnostic
sylva 0.0457 0.0294 0.0434 0.9905 0.994 0.9912 agnostic
Overall   0.1379 0.1973   0.9334 0.8826 prior

This entry is a complete prior knowledge entry.