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.

xent quadratic lssvm

Submitted by reference

LS-SVM with a normalised inhomogenous quadratic kernel. Model selection using scaled conjugate gradient descent, minimising the cross-entropy criterion.

Performance Estimates:

ADA - BER = 0.187897 : AUC = 0.899021

GINA - BER = 0.058147 : AUC = 0.984800

HIVA - BER = 0.281340 : AUC = 0.736672

NOVA - BER = 0.051653 : AUC = 0.986466

SYLVA - BER = 0.011161 : AUC = 0.998760

Dataset Balanced Error Area Under Curve  
Train Valid Test Train Valid Test
ada 0.1682 0.2127 0.1876 0.9165 0.8742 0.8984 agnostic
gina 0 0.0349 0.0509 1 0.9937 0.9884 agnostic
hiva 0 0.2811 0.3169 1 0.7664 0.725 agnostic
nova 0.0004 0.074 0.0488 1 0.9957 0.9901 agnostic
sylva 0.0093 0.0061 0.0121 0.9991 0.9989 0.9984 agnostic
Overall 0.0356 0.1217 0.1232 0.9831 0.9258 0.9201 agnostic