Janelle Shane

Insufficient data may not compute, but it still loves you.

By Janelle Shane On September 29, 2017 - 6 min read

I train neural networks to imitate real-life human things, from fortune cookies [http://lewisandquark.tumblr.com/post/159914835627/fortune-cookies-written-by-neural-network] to to Harry Potter fan fiction [http://lewisandquark.tumblr.com/post/164291045392/harry-potter-and-the-word-level-recurrent-neural] to guinea pig names [http://lewisandquark.tumblr.com/post/162263942282/more-neural-network-guinea-pig-names] . Unlike traditional computer programming where a human

Harry Potter spells, generated by neural network

By Janelle Shane On September 21, 2017 - 3 min read

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,

A neural network learns to create better D&D spells

By Janelle Shane On September 15, 2017 - 3 min read

Neural networks are a type of machine learning program that learns from examples they’re given, rather than relying on a human programmer to invent rules. In an earlier experiment [http://lewisandquark.tumblr.com/post/159363915392/new-dd-magic-spells-designed-by-neural-network] , I trained a neural network to write new names for Dungeons and Dragons

Don’t use a neural network to name your next pub

By Janelle Shane On September 07, 2017 - 3 min read

The King’s Arms. The Bell and Bucket. The Black Bull Inn. The Beggar’s Bridge. A pub’s name is part of its soul, often highly unique, often hundreds of years old, often with a story behind it. A simple computer program couldn’t hope to give a proper

A neural network invents diseases you don’t want to get

By Janelle Shane On August 31, 2017 - 2 min read

Science fiction writers and producers of TV medical dramas: have you ever needed to invent a serious-sounding disease whose symptoms, progression, and cure you can utterly control? Artificial intelligence can help! Blog reader Kate [http://www.mynameiskate.com/blog/] very kindly compiled a list of 3,765 common names for

New My Little Ponies, designed by neural network

By Janelle Shane On August 24, 2017 - 3 min read

The Kingdom of Equestria is inhabited by thousands of colorful, magical ponies, whose life cycle, socioeconomics, and biomechanics are best not investigated too closely. Their names are usually something like “Rainbow Dash” or “Diamond Tiara” or (my favorite because she’s totally a grad student pony): ‘Twilight Sparkle”. Often the

Harry Potter and the word-level recurrent neural network

By Janelle Shane On August 17, 2017 - 6 min read

> Welcome Home by Alisanne | Severus will quite continue in a muggle book. In a previous experiment [http://lewisandquark.tumblr.com/post/162668008357/harry-potter-and-the-neural-network-fan-fiction] , I trained a character-level recurrent neural network (char-rnn [https://github.com/crazydonkey200/tensorflow-char-rnn]) to generate titles and summaries of Harry Potter fan fiction, based on over

Try these neural network-generated recipes at your own risk.

By Janelle Shane On August 06, 2017 - 4 min read

Stuck in a rut in the kitchen? Tired of preparing sandwiches the same old way? Machine learning can help! I trained a neural network [https://github.com/karpathy/char-rnn] on over 30,000 examples of cookbook recipes, and it learned to produce new recipes of its own. You can learn

Craft beer names, invented by neural network

By Janelle Shane On August 03, 2017 - 3 min read

With over 4,000 craft breweries in the United States alone, people are having trouble finding unique names for their beers. That’s a problem, because when two breweries accidentally use the same name, it results in potential confusion at best and at worst, a legal battle [http://www.npr.

Darth Net: Star Wars characters invented by neural network

By Janelle Shane On July 27, 2017 - 2 min read

Neural networks are a type of computer program that imitate the way humans learn. If you give them a long list of examples of anything - paint colors [http://lewisandquark.tumblr.com/post/160985569682/paint-colors-designed-by-neural-network-part-2] , Broadway musicals [http://lewisandquark.tumblr.com/post/162939563362/terrible-broadway-musicals-imagined-by-neural] , cat names [http://lewisandquark.tumblr.com/

Insufficient data may not compute, but it still loves you.

Harry Potter spells, generated by neural network

A neural network learns to create better D&D spells

Don’t use a neural network to name your next pub

A neural network invents diseases you don’t want to get

New My Little Ponies, designed by neural network

Harry Potter and the word-level recurrent neural network

Try these neural network-generated recipes at your own risk.

Craft beer names, invented by neural network

Darth Net: Star Wars characters invented by neural network

Writing

Subscribe

Recent Posts

Botober 2024

Bonus: "Ignore all previous instructions" gets weirder

An exercise in frustration

Bonus: A unicorn goes downhill

Hidden 3D Pictures

Hidden sheep

Follow

Botober 2024

Bonus: "Ignore all previous instructions" gets weirder

An exercise in frustration