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.

Data Grid (CMA)

Submitted by Marc B oulle

Data Grid (CMA)

Full agnostic submission

Data Grids extend the MODL discretization and value grouping methods to the multivariate case (paper submitted to IJCNN 2007).
An ensemble of data grid is averaged according to the compression-based averaging schema (CMA) (see SNB(CMA) submission)

BER guess:
ada: 0.192
gina: 0.147
hiva: 0.310
nova: 0.135
sylva: 0.023

Dataset Balanced Error Area Under Curve  
Train Valid Test Train Valid Test
ada 0.1737 0.1818 0.1761 0.8307 0.821 0.8321 agnostic
gina 0.1289 0.1268 0.1436 0.9334 0.9428 0.9221 agnostic
hiva 0.2651 0.3276 0.3242 0.7659 0.7647 0.717 agnostic
nova 0.1094 0.096 0.1229 0.9226 0.9338 0.9159 agnostic
sylva 0.013 0.0069 0.0158 0.9888 0.9881 0.9873 agnostic
Overall 0.138 0.1478 0.1565 0.8883 0.8901 0.8749 agnostic

This entry is a complete agnostic learning entry.