Have regular Valentine’s gifts got you down? Have candy slogans become too expected? Luckily, research scientist Janelle Shane has our back. Shane fed a whole bunch of candy heart slogans through a neural network learning algorithm, and then was generous enough to share the results with the public. They. Are. Spectacular.
Here’s what happened when a neural net tried to write candy heart messages.https://t.co/tPZyGV3upg pic.twitter.com/4XSkvgEtbb
— Janelle Shane (@JanelleCShane) February 11, 2019
Tag yourself, I’m “book.” Or maybe “time hug.” Or maybe all of them.
But wait, there are more!
Neural net-generated candy hearts range from sweet to questionable.
more here: https://t.co/EFYbvZueUb pic.twitter.com/PVXEmofj2G
— Janelle Shane (@JanelleCShane) February 11, 2019
Eventually, the neural network got a little carried away with the *ahem* romance.
so, um. neural net generated these Highly Questionable candy hearts
some of these words were NOT in its training data, but it was technically allowed to use any letters in any order. pic.twitter.com/9dTfDalyJG
— Janelle Shane (@JanelleCShane) February 12, 2019
Let us know which possibly nonsensical/potentially inappropriate candy heart you’d send to your bae/self (i.e. the ultimate bae). (via AIWeirdness)
- This is a really great conversation between two paranormal romance writers about what makes that genre so very cool. (via A Love So True)
- This MSNBC graphic was already garbage before I noticed that they accidentally used Sherrod Brown’s picture twice! Not even major media outlets can tell these many white men apart.
*drags file to trash* pic.twitter.com/yEHZJ04Chu
— Lauren Duca (@laurenduca) February 14, 2019
- Paddington is headed to the small screen! (via Indiewire)
Happy Valentine’s/regular Thursday, everyone!
(image: screencap)
Want more stories like this? Become a subscriber and support the site!
—The Mary Sue has a strict comment policy that forbids, but is not limited to, personal insults toward anyone, hate speech, and trolling.—
Published: Feb 14, 2019 06:10 pm