Artificial Neural Network Writes a Full House Episode Everyday, Forever

A never ending nightmare


Mon, May 2nd, 2016 11:00 by capnasty NEWS

Possibly in response to Netflix's launch of Fuller House, Sam Kronick and Joe Veix wrote a program that, using a artificial neural network machine learning algorithm, writes a new episode of Full House everyday, for all eternity. Joe explains how it works here.

To generate text, this particular RNN is programmed to read an input (for example, Full House scripts), then generate an output based on patterns it notices. It compares the new file with the original, noting the hits and misses, and adjusts itself accordingly (hence: machine learning), before creating yet another new version. It repeats this process over and over, thousands of times, until it starts creating fairly accurate approximations of the original. Unlike Markov chains, which work on a word-by-word basis with no memory, RNNs can work on a character level, actively learn, and consider fancy things like paragraph structure and formatting.



You may also be interested in:

The Binary Low Table: Furniture Made Out of Obsolete Computer Parts
Vintage Terminal: Terminal Emulator Emulates 1980s Monitor
Couch Multitasking No Longer Just for Nerds
Los Alamos National Labs Has Been Running a Quantum Internet for 2 Years
Moonjs: the Apollo Guidance Computer Simulated in Javascript