A machine named Giraffe taught itself to play chess and reached a level of skill on par with master chess players in just three days, according to The Independent. Matthew Lai, the Cornell University researcher who built the machine, implemented a “neural network” into it to allow it to utilize information and situations from real games of chess and simulate human thought-processes, although it was never explicitly told the rules.
The networks that exist in the human brain inspired the “neural network” design; each time the nodes of the network receive information, they change and adapt just like the human brain does during the process of learning.
Before the existence of Giraffe, Deep Blue was the other well-known computer that could play chess, according to Fusion. However, whereas Deep Blue used a brute-force method of scanning every potential move available until it found the best one, Giraffe mimics the decision-making process that humans use.
“Unlike most chess engines in existence today, Giraffe derives its playing strength…from being able to evaluate tricky positions accurately, and understanding complicated positional concepts that are intuitive to humans, but have been elusive to chess engines for a long time,” said Lai in his report.
After three days, Giraffe was able to play at a level that placed it within the top 2.2 percent of professional chess players, according to MIT Technology Review.
Despite Giraffe’s success at chess, it was created primarily to showcase the abilities of computer learning and its potential to surpass that of humans.