(un)stable equilibrium

(un)stable equilibrium (2019) is an ongoing series of works that are borne out of an artistic practice developed around the training of generative neural networks without any training data.

In the first series of works, the arrangement of networks is similar the Generative Adversarial Networks architecture. But for these works we have two generators that try to have their output passed of as the other network’s, while both competing to have more diversity in the colours they produce compared to the other network.

The piece (un)stable equilibrium 1:1 won the Grand prize at the ICCV 2019 Computer Vision Art Gallery. The work was presented as a poster at the NeurIPS 2019 Workshop on Machine Learning for Creativity and Design.