Convolution

BACKGROUND Contemporary artificial intelligence primarily operates via a representational model, which makes it problematic for incorporating artificial intelligence into interactive artworks, since there is an expectation of reciprocal engagement. The scale of this problem in the art world is evidenced by the ongoing discussion by art critics like Johnathon Jones, and significant ongoing attempts by contemporary art galleries like the Tate Britain to countenance this problem. Similarly, contemporary robots are problematic in terms of agency within an artwork, as explored in the work of Petra Gemeinboeck and Rob Saunders, for example. CONTRIBUTION This work addresses these problems by developing a model of 'artificial desire' in an attempt to introduce a non-representational agency to the roles of both artificial intelligence and robots within the work. 'Convolution' is an interactive installation that lets human participants and a child robot co-create an immersive audiovisual artwork through the use of the robot's artificial neural networks to enable object and image recognition, which is then evolved via 'artificial desire' algorithms. The resulting artwork dissolves the boundaries between computational and physical phenomena, displaying an aesthetic that is a real hybrid of the physical and the digital, of human and machine learning, of natural and artificial intelligence, and of real and synthetic evolution. It is an artwork and aesthetic that emerges from the interaction between robot, people, deep learning AI and virtual environment, neither one taking precedence, rather collaborating on a genuinely post-digital, post-convergent artwork. SIGNIFICANCE The quality and significance of the work is attested to by its selection into the SIGGRPAH Asia 2016 Art Program, one of the world's most significant conference and exhibition on computer graphics and interactive technologies.