Code, computer, self-learning algorithmic drawing, vector graphics, 65” screen.
Pachinko Machine is a self-learning algorithmic drawing constructed as a pinball machine. The Machine plays the game a game of pinball itself in an automated setup. During the period of the exhibition the Machine’s algorithm optimizes its own performance through machine learning and becomes increasingly good at trying to win the game of pinball. The only barrier to success is beyond the algorithm’s control: there is a second intelligence embedded within the machine that aims to obstruct and obfuscate. The work is a battle of algorithms.
The visual part of the work is created in a computer drawing process that uses only cycles and triangles to create complex visual appearances. The drawing process is developed in a systematic manner from the top to the bottom without the artist’s comprehension what the final image may look like. The creation of the drawing takes months of drawing, erasing, recreating and fine-tuning. The machine comes full to life when kinetic attributes are assigned to each element.
Pachinko Machine reflects on the current discourses around artificial intelligence and self-thinking system. Machine learning algorithms are poor in simulating human decision-making; they are however rather successful at understanding and playing games created by humans (for example Deep Mind’s recently developed algorithms). Pachinko Machine takes one step further and humanizes the machine learning process by introducing chance into the game. The battle of these algorithms that aspire to disparate goals aestheticises the process driven by machine learning. Viewers are welcome to make sense of this battle whilst observing the mesmerising patterns of the Pachinko’s drawings.
The work aims to represent the walk of life and although we may be able to control some part it, there is always an element of chance that may destruct and encourages us to stray from our original path. Pachinko Machine was developed in Unity. brigittazics.com/work/pachinko-machine
- Brigitta Zics is an award-winning artist who creates works with visual and material sensitivity that seeks to fashion new experiences. She works on the convergence of art and science and explores mixed-media forms combining various techniques and emerging technologies. Her recent interests embrace experiential art, human perception and art, and the aesthetics of data and algorithms. Her works has been exhibited at Budapest, Berlin, London, Los Angeles, Monaco and Cologne and published at Leonardo Journal (MIT press) and other academic journals (Springer, Taylor &Francis). In 2004 her work was nominated for the International Media Art Award at ZKM, in 2005 she was selected for one of Europe Best Multimedia Artists. She is an Associate Professor at Ravensboure University of London and Fellow of Royal Society of Arts, UK. brigittazics.com