N. Matic, I. Guyon, J. Denker, and V. Vapnik (1993)
Writer adaptation for on-line handwritten character recognition.
In: In Second International Conference on Pattern Recognition and Document Analysis, ed. by IEEE Computer Society Press, pp. 187–191, Tsukuba, Japan.
A TDNN without its last layer serves as a preprocessor to an SVM classifier, that is retrained to peculiar writing styles. This combination allows for fast writer dependent learning of new letters and symbols.