Last week
[http://lewisandquark.tumblr.com/post/165373096197/a-neural-network-learns-to-create-better-dd]
, I trained a neural network [https://github.com/karpathy/char-rnn] to generate
the names of magic spells from the game Dungeons and Dragons, by giving it a
list of existing spells and asking it to generate more like them. It obliged,
![Harry Potter spells, generated by neural network](/content/images/size/w795/image/fetch/w_1200-c_limit-f_jpg-q_auto:good-fl_progressive:steep/https-3A-2F-2Fbucketeer-e05bbc84-baa3-437e-9518-adb32be77984.s3.amazonaws.com-2Fpublic-2Fimages-2Fa2c7c912-2468-4adf-b37c-57d1339a2153_489x489.png)