Artist Statement
This exhibition is focused on using artificial intelligence as a catalyst for abstract 3D modeling. Art has also been deemed as something that only a human can do, and that a machine or computer is incapable of. This exhibit is here to challenge that notion. Using a machine learning method called Generative Adversarial Networks (GAN) I have learned what it takes to train a computer to generate unique images and 3D models in record time. A GAN is like that of how a human brain learns and stores information. A major part of the exhibit is understanding this psychology of the artificial intelligence and understanding what it needs to be able to generate models successfully. After 6,500,000 unique images generated, thousands of training hours, and many more troubleshooting, I present the results of my research and process. This exhibit should not be seen as the limits of this technique, but only the beginning. I hope that this exhibit will inspire someone to take up the field of machine-based creativity and advance our understanding of the relationship between art and computers.
![](https://i0.wp.com/sites.uwm.edu/studio-art-ba-bfa/files/2021/05/robert_polivka_1.jpg?ssl=1)
![](https://i0.wp.com/sites.uwm.edu/studio-art-ba-bfa/files/2021/05/robert_polivka_2.jpg?ssl=1)
![](https://i1.wp.com/sites.uwm.edu/studio-art-ba-bfa/files/2021/05/robert_polivka_3.jpg?ssl=1)
![](https://i1.wp.com/sites.uwm.edu/studio-art-ba-bfa/files/2021/05/robert_polivka_4.jpg?ssl=1)
![](https://i0.wp.com/sites.uwm.edu/studio-art-ba-bfa/files/2021/05/robert_polivka_5.jpg?ssl=1)