A deep learning visualization of your own memories
By Aubrey Simonson
Commissioned by Manuel Cebrian and Iyad Rahwan
Nostalgia Box is a continually shifting soup of memories. Images, curated for their nostalgic emotional impact, are dreamed over one another using a neural style machine learning algorithm, based on the paper "A Neural Algorithm of Artistic Style" by Leon A. Gatys, Alexander S. Ecker, and Matthias Bethge. The resulting images are then overlaid into a video which never fully solidifies into any one image, but instead is always several at once. To someone who is familiar with the content of the images, they are still recognizable. However, a stranger should see only a haze of vague shapes which occasionally contains the suggestion of a face. This is machine learning, stripped of the elements of spam. It is an amalgamation of its creator's digital history, but which is being used as a tool for reflection, rather than as a means to more effectively market products.