AI reimagined The Great British Bakeoff as a weird acid trip, in case you're hungry for hoverbread or void cake
A neural net shows off its idea of a perfectly terrifying baking show I wish really existed.
The Great British Bake Off -- rebranded for Americans as The Great British Baking Show on Netflix -- features amateur bakers competing to make the best cakes, breads and other delectable goods inside a large tent out in the country. But when artificial intelligence tries to understand the charming popular baking show, things start to look like a horror film with desserts.
In her latest experiment, AI researcher Janelle Shane trained a neural network to use 55,000 screenshots from the show featuring images of bakers, pastries, tents and even random wildlife. The neural net spit back what it thought best represented the quaint baking show, and the results aren't tasty.
The Bakeoff neural net is so much better at periodic structures that sometimes it will try to make HUMANS be periodic structures.
— Janelle Shane (@JanelleCShane) March 27, 2020
they are not, neural net.
they are not. pic.twitter.com/WtTXiRGsfu
Shane used a state-of-the-art image-generating neural net called StyleGAN2 that's rather good at understanding the concept of human faces. But it doesn't do so well when you add in human bodies, cakes, pastries and tents.
The first thing the neural net did after it was fed all the images was erase the human faces from The Great British Bakeoff screenshots. Apparently, the neural net was not only confused by human faces that weren't positioned dead center in the screenshots, it was also having difficulty figuring out external shapes of baked goods and the interior of the tent.
Trained a neural net on the Great British Bakeoff
— Janelle Shane (@JanelleCShane) March 27, 2020
results were less than cozy
it tried, though
StyleGAN2 via @runwayml https://t.co/u8g5lnlmaR pic.twitter.com/khMTJzhqzN
Neural nets are great at understanding patterns, so when presented with an image of one pie or one human body, it wants to replicate it over and over in the same screenshot to show off a pattern. That means human bodies in a screenshot might end up with extra arms.
"A neural net usually builds images by stacking lots of repeating features on top of one another, fine-tuning the balance between them to produce objects and textures," Shane wrote in her blog. "If it gets the balance slightly wrong, individual repeating features tend to pop out."
this seems to be the neural net's consistent rendition of "baked goods"
— Janelle Shane (@JanelleCShane) March 20, 2020
i'm not sure it's safe to eat pic.twitter.com/Qtjwlp32QH
The most amusing part of this experiment was seeing what the neural net thought was the ideal baked good. Some of the more memorable items include a cake with weird holes, hoverbread and a blueberry pie with way too many layers. Yum!
Shane's previous neural net food experiments have proven to be both bizarre and entertaining. Shane trained her neural network to come up with weird Harry Potter pie creations, unusual cookie names and Valentine's Day candy heart sayings.
You should read it
- 91% of technology managers believe that AI will be the center of the next technology revolution
- [Infographic] Benefits and hazards from Artificial Intelligence
- Watching pictures painted by artificial intelligence, everyone thinks that is the work of a true artist
- Top 5 programming languages to develop AI
- Artificial intelligence learns to create another artificial intelligence, replacing people in the future
- 6 steps to start learning artificial intelligence programming (AI)
- Now, artificial intelligence can predict how long you will live
- What happens if aliens are artificial intelligence?
- How does Al revolution help us answer the most fundamental problems of philosophy? (Part 3)
- How does Al revolution help us answer the most fundamental problems of philosophy? (Part 2)
- How does Al revolution help us answer the most fundamental problems of philosophy? (Part 1)
- How can the AI see us behind the walls?
Maybe you are interested
The killer bird blinds and defeats poisonous African snakes Why do shrimp and crab shells turn red after cooking? The Japanese can make flying cars, just like high-speed maglev trains What happens when lightning strikes a lake? Is it good to eat guava, including the skin? 10 strangest military vehicles in world history