Generative Major Arcana


Generative Major Arcana is both a technical and a conceptual piece that explores the use of a DCGAN (Deep Convolutional Generative Adversarial Network) algorithm to generate Tarot symbols, with the question: is meaning inherent to symbols or is merely assigned by us?. Oracle decks for divination –Tarot among them– are symbolic frameworks built during a process of cumulative cultural development, so can be decoded in different contexts and cultures. On the other hand, Machine Learning can be conceptually simplified to algorithms that analyze certain data in order to find patterns and then reproduce them. 

Then, new symbols –apparently with no meaning– are generated based on well known ones and it depends on us and our bias –and Machine Learning bias– to decode these symbols and assign a significance.

This slideshow requires JavaScript.