Research scientist Janelle Shane has been using neural networks to come up with new paint names, and they're pretty hilarious.
It's hard having to come up with new names for everything. Think about how many names their are for new products out there, it's difficult coming up with something genuinely unique. So why not solve the problem by using Artificial Intelligence? This is exactly what research scientist Janelle Shane has done, yet she decided to create new paint names through AI.
In an interview with Gizmodo, Shane admitted that she was only really doing this for fun. “What inspired me was I found a post online from someone who’d done neural network cookbook recipes,” she revealed. "I thought they were hilarious and I wanted more, but there weren’t any more. The only way to get more was to make more.”
[caption id="attachment_7862" align="aligncenter" width="282"] "Burf Pink" sounds like a nice colour...[/caption]
Shane used neural networks to come up with her new and brilliant paint names. These are computer systems that are loosely based around the way our own brains work. They can be trained on large datasets, and usually used in areas such as Natural Language Processing and image recognition.
In order to come up with these new paint names, Shane used fed a list of 7700 paint colours from Sherwin Williams into a neural network called char-rnn. This resulted in getting new paint names such as "Sindis Poop" and "Stanky Bean". What a world we live in. Have a look at some of the other names below.
[caption id="attachment_7863" align="aligncenter" width="281"] I will cover my walls in "Stanky Bean".[/caption]
During the neural network's learning process, Shane took some early names from the system, and of course they were complete nonsense. Yet eventually she started seeing actual words being produced and the final results were wildly entertaining.
“It’s tempting to correct the spelling if it almost spells a word, but somehow that takes the fun out of it,” said Shane told Gizmodo. “This is as it comes out of the computer, I’m not changing a thing.”
“I’m sure ‘Hurky’ wasn’t in the original dataset,” commented Shane. “But somehow it’s come up with that.” Shane is just doing this for her own amusement, but you can use char-rnn for yourselves and end up with some hilarious results.