everyone has that one thing they were obsessed with learning about as a kid that never goes away like it could have been years since you last looked something up related to it but the passion is still there man
do me a favour and reblog with what your childhood obsession was like I am so curious about everyone else’s because it can be the most specific thing and it’s amazing
So if you’ve ever picked out paint, you know that every infinitesimally different shade of blue, beige, and gray has its own descriptive, attractive name. Tuscan sunrise, blushing pear, Tradewind, etc… There are in fact people who invent these names for a living. But given that the human eye can see millions of distinct colors, sooner or later we’re going to run out of good names. Can AI help?
For this experiment, I gave the neural network a list of about 7,700 Sherwin-Williams paint colors along with their RGB values. (RGB = red, green, and blue color values) Could the neural network learn to invent new paint colors and give them attractive names?
One way I have of checking on the neural network’s progress during training is to ask it to produce some output using the lowest-creativity setting. Then the neural network plays it safe, and we can get an idea of what it has learned for sure.
By the first checkpoint, the neural network has learned to produce valid RGB values - these are colors, all right, and you could technically paint your walls with them. It’s a little farther behind the curve on the names, although it does seem to be attempting a combination of the colors brown, blue, and gray.
By the second checkpoint, the neural network can properly spell green and gray. It doesn’t seem to actually know what color they are, however.
Let’s check in with what the more-creative setting is producing.
…oh, okay.
Later in the training process, the neural network is about as well-trained as it’s going to be (perhaps with different parameters, it could have done a bit better - a lot of neural network training involves choosing the right training parameters). By this point, it’s able to figure out some of the basic colors, like white, red, and grey:
Although not reliably.
In fact, looking at the neural network’s output as a whole, it is evident that:
The neural network really likes brown, beige, and grey.
The neural network has really really bad ideas for paint names.
The crows she feeds obviously have their own little lives. They go about their business, and they spot *pretty thing* or /unique thing/ in question. What gets me is that the *first* thing on their minds as recipient of this thing is the little girl that feeds them.
They spot a thing, and immediately must think, “that nice girl with delicious foodstuffs must have this to show my gratitude.”
It’s actually more than that, though, if you read the articles or watch the videos. This has taken place over YEARS- it started with these birds following this little girl around because she was a messy eater and it has turned into a ritual for the family. They have a water station and food stations where they daily set out things for these birds and sometimes (but not always), these birds leave ‘payment’ behind for the food.
BUT WAIT THERE’S MORE
These birds are not just taking food and leaving shinies. These birds are watching over this family now. Their lives have become involved. These crows are keeping track of this girl and her mother even when they are out of the yard. How do we know?
One of them is a photographer, and one day while she was photographing some stuff on a bridge, she dropped her camera’s lenscap over the edge. There was no way she could get it back, so she left it. When she got home, the lenscap was sitting on the edge of one of the feeding stations, waiting for her.
Not only were the birds following and watching over her, they were smart enough to realize she dropped an Important Thing and cared enough to bring it back to her.
when i see a thing that says “this is the funniest thing i’ve ever seen” i generally prepare to gently chortle but instead while i was watching it my partner messaged me that “you sound like a ghost”