ConceptAsymptotic Error
Asymptotic Error
The generalization error to which a student network converges as the number of training samples tends to infinity, reflecting optimal performance under unlimited data. It is determined by model misspecification and optimization bias.
Timeline (last 90 days)
Oct 22
Spatial correlation reduces the advantage of weight sharing: the asymptotic error of the convolutional student rises as the input spatial correlation c increases, whereas the asymptotic error of the locally connected student is independent of c and is overall lower at the same learning rate.
Unverified50%