Experimenting with Machine Learning

Currently, we are at a pivotal period in history. Facing a global pandemic, living through times like never known before. Since the start of 2020, we have been flooded with bad news and sad stories regarding COVID-19. We need to be positive in these difficult times by focusing on the number of newborns, newly pregnant women, fighters and survivors to create a portrait of COVID-19.

The daily statistics during the period when this research is being conducted suggest that about 116million babies will be born in the shadow of COVID-19, while there are approximately 12 milliondeaths around the world. It is important to illuminate a positive narrative about what is hidden beyond COVID-19, and through the medium of art, we can feel in a way that we cannot feel through statistics and numbers.

In quarantine, we have no choice but to meditate. This means emptying a mind filled with thoughts in order to rethink deeply about ourselves. It is a vital goal to experience mindfulness objectively. Although art is not an essential medication, it is nevertheless essential for human beings. Throughout history, artists have mirrored, discussed, challenged and analysed the world in which they live and the times which they are going through.

As an artist, I am curious about how I can respond to this pandemic in a way that might change my artistic practices, particularly after launching my solo virtual show and website. I want to create an experience in machine learning as a way of using artificial intelligence. In this work, I want to draw a positive narrative of the pandemic.

Generated images by the artist through experimenting training image in ML

My work is a kind of portrait drawn by software named RunwayML. Collecting medical imaging of newborns and the newly pregnant to create a portrait of COVID-19. It is going to focus on the population growth from millions of pictures, the medical images of pregnant women and newborns in 2020. I shall try to collect data from the World Health Organization and UNICEF in as much accurate detail as I can in order to move the viewer to a distant and abstract place. I want to convey the beauty of our body through medical imaging to inspire viewers to think critically about the hidden face of Corona.

I am starting to use two of the resources of ML such as Runway ML: An easy, code-free tool that makes it simple to experiment with machine learning models in creative ways. Our overall staff pick, and GANBreeder – Breed two images to create novel new ones using GANBreeder. (Note that GANbreeder was renamed ArtBreeder, with several AI models to manipulate photos).