The Schaubies' Story |||

Boxes, Boxes Everywhere

Personal Log by Fred, 30.07.2020

This week I spent one afternoon drawing boxes around the whales that I saw in 200 pictures of whales and around black holes in around 200 pictures of black holes. It made me understand that a computer can really only see“ what I teach it. If I don’t teach the computer that whales exist, then there are no whales in its word. Furthermore the computer learns from humans where things start and where things end (where does a black hole end and where does it start??).
So: Based on the things we see, we teach an A.I. to get it right. This made me very very sad and I do not really know why.

Week #3 was tough for me. It was a lot about getting things right: finding the right codes, the right pictures, the right way to classify pictures. While Robert programmed his fingers off I tried to teach myself the basics of how neural networks work - I also wanted to get it right!

pride (Photo by )

Last Saturday was Cristopher Street Day in Berlin. It was a beautiful hot summer day and it was thrilling to walk the streets with my little sign Trans* Rights are Human Righs“ made from a cardboard box, surrounded by people who shared the same (/similar) queer views as I do.

While trying to get things right this week, it occurred to me, that all this image recognition stuff is really contradictory to what trans* people are fighting for. Is it not the exact same thing that oppresses trans* rights? That someone once recognized certain features about a body and based on this put it in a little box saying male“ and female“. Could we make an image recognition A.I. that respects the wishes of the detected objects?

Until then, if you are in Germany, please consider signing this petition about changing the law: replacing the Transsexuellengesetz“ with a new law that will allow people to decide themselves what gender they are:[]

Thank you.

Up next Connecting the dots the technology of collaborative work, and other feelings
Latest posts The Last Schaubie Chain story Live with a computer vision software Baby steps with our own vision AI Telling worlds the technology of collaborative work, and other feelings Boxes, Boxes Everywhere Connecting the dots do sheep dream of fluffy robots? The cat and the musician week#3 games x3 21282 sci-fi images (for neural-network training) Narration Image Description and Restaging Trying the google vision API with drawings day#5 gcloud Vision API Example day#4 day#3 flag day#2 day#1