Logistic regression over 10 bootstrap samples.
Trained for 30 epochs with 0.1 learning rate, 0.5 momentum.
Tested on out-of-sample examples for each bootstrap.

N1: All non-positive genes are assumed negative.
N2: Direct annotations to any ancestor node are not used for negatives.

AUCs: Mean ± StDev of Area under ROC Curve over 10 bootstrap samples.
C.Mean Area under ROC Curve of combined classifier (Bagging), aggregating predictions by mean (regression).
C.Median Area under ROC Curve of combined classifier (Bagging), aggregating predictions by median (voting).
Each Bagging weak classifier only contributes test output on examples held-out from it.

Click column headers to sort!