MACHINE LEARNING CURATION
For this series of original paintings, I have been painting abstract compositions by hand. I am researching machine learning as a technology to curate and classify painting compositions. I use machine learning software I wrote & supervise to extract shapes from my existing abstract paintings, create new compositions and then suggest which ones I will like most. I’ve digitally created over 40,000 compositions that the software aids me in ranking & suggesting my preferred compositions. I have testing different classification processes while developing new software including: Linear Regression, Random Forests, Decision Trees, K-Means clusters, and TensorFlow. Although I am using machine learning, the most facinating aspect of the project for me is how much I am learning about subjective preferences and abstract composition.
It has also been interesting documenting the learning process through the act of painting. Painting is seen as a valuable, resource intensive process, while software development is quick. I am spending resources painting results of instantaneous calculations that are in development.
The paintings are titled based on Google’s untrained TensorFlow Inception model. Untrained means it is downloaded from Google and that I have not trained the model for my specific task. Their model classifies images to objects, so the software suggest what the abstract painting is. It is interesting to take a second look at abstract art and consider how its been seen by the code. For example, hockey puck, bird house, sweet potatoe, bullet proof vest are only some of the objects it has found in the paintings.
I will be showcasing these artworks at the 2018 East Vancouver Culture Crawl on November 15-18, 2018 in my open studio. Here is a selection of the new artwork: