These are all costumes generated past times a neural network trained past times Jacobs School alumna Janelle Shane. Shane, who earned a Ph.D. inward electrical applied scientific discipline at UC San Diego inward the lab of Professor Shaya Fainman, industrial plant amongst lasers past times day. But her hobby is working amongst neural networks to practise funny information sets.
For this project, she crowdsourced 4500 costume ideas from her weblog readers as well as fed them to a neural network.
The network did non disappoint, generating costume suggestions such every bit vampire Big Bird, celery bluish Frankenstein as well as strawberry shark.
Soon, Shane's readers were getting into the game as well as drawing the costumes the neural network suggested.
First up, strawberry shark:
And then: Bearley Quinn (courtesy of Twitter user @vonbees):
But Shane's readers weren't done. Soon they started making simply about of the costume suggestions a reality.
Twitter user Liz Walsh dressed upwards every bit the Dragon of Liberty:
Twitter user @HerbLovesTech and his married adult woman dressed upwards every bit Professor Panda as well as Shark Princess:
And Shane? She took her inspiration from an entry inward the costume information base. She volition live Ruth Vader Ginsburg (that's a squelch upwards of Supreme Court Justice Ruth Bader Ginsburg as well as Star Wars villain Darth Vader):
For to a greater extent than neural-network generated Halloween costumes, read Shane's weblog post here. And read this news floor past times author Rae Paoletta here and this Popular Mechanics story past times writer Sophie Weiner. Sumber http://jacobsschoolofengineering.blogspot.com/