“Guess, check and fix: a phenomenology of improvisation in ‘neural’ painting” is now published in the journal Digital Creativity. You can access the paper here.
The parameter space offered by neural network image synthesis offers a creative environment that is little understood and quite literally emergent. In an attempt to come to terms with this space, an artist enters into an improvisational reflection on ‘neural painting’, made possible with what are called style transfer algorithms (Gatys, Leon A., Alexander S. Ecker, and Matthias Bethge. 2015. “A Neural Algorithm of Artistic Style.” ArXiv: 1508.06576 [Cs, q-Bio], August. http://arxiv.org/abs/1508.06576)Artistic painting offers access to transitional states existing at the interstices of expression and reflection in the creative process. Of all the conceptual dimensions offered by neural style transfer models (where the ‘content’ of one source is blended with the ‘style’ of another), the convolutional blending of ‘content weight’ offers a fertile metaphor for artistic painting phenomenology, providing a tool for the investigation of stylistic schema in the iterative, improvisational movement from concept to representation. A preliminary phenomenological framework describing the process of neural painting is developed, offering an art-as-research perspective on intersubjectively positioned creativity support technology.
KEYWORDS: Art process, embodied cognition, reflective practice, AI art, J. G. Ballard