Baby names generated by a neural network

Single-layer_feedforward_artificial_neural_networkAndrej Karpath, a Stanford computer science PhD candidate, designed a plausible baby name generator, while working on recurrent neural networks (a type of artificial neural network).

According to Wikipedia, “in machine learning and cognitive science, artificial neural networks (ANNs) are a family of models inspired by biological neural networks (the central nervous systems of animals, in particular the brain) which are used to estimate or approximate functions that can depend on a large number of inputs and are generally unknown.”

In this case, Karpath fed the neural network 8000 real baby names as input and generated plausible baby names not in the original data set.

Examples: Antley, Nerille, Chelon, Walmor, Evena, Jeryly, Stachon, Charisa