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.

out5-valid-no-fs (Intel final 3)

Submitted by IDEAL, Intel

out5-valid-no-fs submitted on 0707310214

Dataset Balanced Error Area Under Curve  
Train Valid Test Train Valid Test
ada 0.1863 0.2078 0.1856 0.8133 0.796 0.8145 agnostic
gina 0 0 0.0556 1 1 0.9441 agnostic
hiva 0.2454 0.2792 0.3058 0.7542 0.7403 0.6953 agnostic
nova 0.0058 0 0.0584 0.9924 1 0.9423 agnostic
sylva 0.0073 0.0065 0.0136 0.9933 0.9933 0.9864 agnostic
Overall 0.089 0.0987 0.1238 0.9106 0.9059 0.8765 agnostic

This entry is a complete agnostic learning entry.