Blade Runner—Autoencoded is a film made by training an autoencoder—a type of generative neural network—to recreate frames from the 1982 film Blade Runner. The Autoencoder learns to model all frames by trying to copy them through a very narrow information bottleneck, being optimized to create images that are as similar as possible to the original images.
This series of scientific experiments conducted by Dr. Manuela Macedonia and her staff at Johannes Kepler University Linz in cooperation with the Ars Electronica Center and the Catholic University of the Sacred Heart, Milan, investigates learning a language in a virtual setting.
There is a surprising similarity in the way neural networks and analogue synthesizers work: both receive signals and process them through components to generate data or sound. cellF combines these two systems.