Spatial and Time Domain Feature of ERP Speller System Extracted via Convolutional Neural Network

Figure 3

Learning curve and receiver operating characteristic curve (ROC) of L and H subject. (a) False negative rate (FN) and learning curve of L subject saturates near 0 and .2, respectively, whereas false positive rate (FP) increase to 1. (b) Both FP and FN drop over the time course for H subject and learning curve saturates near .8. (c) Training and validation error of drops over the time course for both L subject and (d) H subject. Both validation and training error are lower for H subject. (e) ROC curve of H and L subjects.

(a)

(b)

(c)

(d)

(e)