Google AlphaGo Unveils Breakthrough in AI Tech

By Tamanna Desai,
Tech & Innovation Assistant Editor

Google and Facebook are arguably the biggest data collecting sources of our era. Recently, Google had a breakthrough with computer program called “AlphaGo”. The system plays the Chinese game called “Go” and is able to beat any human player, any other computing system and most importantly it is trained to beat itself. Facebook’s version of this kind of software is called “Darkforest”. CEO Mark Zuckerberg posted on his page that he and his team were in the works of coming close to a breakthrough with the AI technology. Facebook’s version of “AlphaGo” is called Darkforest. Artificial intelligence systems started back in the 1990s when programmers taught the computer how to play tic-tac-toe, checkers and chess.

The ancient Chinese game was created about 2,500 years ago according to Google’s blog post on their technology. The Chinese game “Go” is very similar to the game of chess in that there are two players one has black stones, and one has white stones, and they each take turns. The purpose of the game is to take control over the entire board by boxing in your opponent and surrounding them with all of your colored stones. Although it seems very easy for a computer program to play this game, it is much harder than it looks because of the billion variations that could take place. Google’s vision was to teach the system to think for itself- i.e. artificial intelligence. Their other vision is for the program to recognize faces and voices. According to the Guardian, AlphaGo has “intuition” and therefore is not able to explain why it picks certain moves.

The reason why AlphaGo being able to conquer the mission of playing this game and winning every time, is a breakthrough because if the system can figure out strategies on its own, then in the future there is a possibility that we could use these systems to help us out with huge issues that our country is dealing with today. The idea of a software essentially telling us solutions to problems is very intriguing but scary at the same time. Technically the software could outwit us and its cognitive abilities could be incomparable to human cognitive abilities. This could cause problems in terms of how we will be able to control the software. In addition, there is a possibility that the software could provide a solution that is wrong and backfires. On the bright side, the technology could really help us solve medical issues at a faster pace.

This also leads to the issue of open sourcing. Google open sourced some of its algorithms for “Go” which just means it shared some of its tools to the world. Open sourcing helps foster innovation. This was a big shift for Google because the company is founded on their software. It is the equivalent to giving away a recipe that made a certain item the best out of all of its competitors.

From Mar. 9 to Mar. 15, Google’s “AlphaGo” will compete in a tournament with Lee Sedol who is considered the Go game guru of the world. The tournament will take place in South Korea and the winning prize is one million dollars. Creating “AlphaGo” that has a 99.8 rate of winning is a huge development in the computer science and technology world.

 

A version of this article appeared in the Tuesday, February 9th print edition.


Contact Tamanna at
tamanna.desai@student.shu.edu

Advertisements

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s