Here are the most memorable Game of Thrones characters, base
Source: Sara Castellanos
As the fifth season of HBO's Game of Thrones comes to a close, MIT researchers have developed an algorithm that can predict the top 10 most memorable characters in the show.
The list begins with Daenerys Targaryen and ends with Joffrey Baratheon.
The research uses techniques in “deep-learning" ― an emerging field of computer science that teaches computers how to learn tasks and find patterns on their own, according to MIT's Computer Science and Artificial Intelligence Lab.
With the “Game of Thrones” season finale coming up Sunday, fans have been feverishly discussing the show and its many polarizing characters. Who’s the meanest? The sexiest? The most memorable?
For that last one, MIT scientists are on the case.
Researchers from MIT’s Computer Science and Artificial Intelligence Lab (CSAIL) have crunched the numbers on GOT characters using a special algorithm they developed that can predict how memorable an image will be.
Check out the photo gallery above to count down Westeros’ most memorable personalities. (Surprise, surprise: thin, attractive blondes capture people’s attention.)
You can also upload your own photos to see how memorable you and your friends are!
The CSAIL researchers developed a collection of over 60,000 images that represents the world’s largest existing memorability dataset. Each image had received a “memorability score” based on human subjects’ abilities to recall them in a memory test.
The team then fed the images to its software, which analyzed the information to pick subtle trends about the features of these images that made them more or less memorable.
“As humans, we tend to remember and forget the same pictures and faces as each other,, which suggests that memorable images have features that automatically make them easier to remember,” says CSAIL PhD student Aditya Khosla, who developed the dataset with CSAIL graduate student Akhil Raju and MIT professors Antonio Torralba and Aude Oliva.
The research uses techniques in “deep-learning,” an emerging field of computer science that teaches computers how to learn tasks and find patterns on their own. In recent years companies like Facebook and Google have spent hundreds of millions of dollars to acquire start-ups with expertise in the topic.
| }
|