Refactored Landscapes is a show that explores the learning process and the creative output of Artificial Intelligence and Machine Learning through computer-generated imagery and video. Referencing historically Canadian artworks and their aesthetics, this show focuses on the phases and iterative steps in which computational creativity evolves and develops. By displaying these iterations, it might make it possible to glimpse into the workings of the algorithms that many of the platforms we depend on everyday use. The creative learnings or creative evolutionary development that algorithms execute to understand and predict our desires are what makes it possible for online personalized experiences.
Using The Group of Seven as the basis for content and creative inspiration, the intention is to connect and merge an iconic Canadian aesthetic from a specific period in Canada’s past with the present cultural and technological forefront which Canada is currently paving in the Artificial Intelligence and Machine Learning fields.
The three series of work being shown in Refactored Landscapes utilize various methods of Machine Learning to try and share the otherwise hidden world of algorithmic decision making and its evolutionary learnings with the public. The first series uses a machine learning method to learn an artistic style from a source image and then has the algorithm “re-paint” a new content image with the learned artistic style. The second explorative series uses an algorithm to learn an artistic style referencing roughly 100 different style examples. In this case, many Lawren Harris paintings were referenced. The algorithm is then tasked to generate entirely new creative imagery based on the collection of the styled reference images. Finally, the last piece is a visual manifestation of algorithmic creativity that evolves through a linear timeline.
By exposing the iterations and some of the methods used by computers to learn and generate their own artistic style through algorithms, does it alter our acceptance or awareness of how these systems influence how we consume and digest digital content? Or when algorithms create, who is the author or artist? Is it the original artist of the source images, the artist or programmer of the final work, or is it the algorithm itself?
Work from this series is available in the store.