Artificial Intelligence Beats Humans at their Own Games
Gamers are a competitive bunch who have just met their match - an artificial intelligence program called AlphaStar.
Some of the world's best players of the real-time strategy game Starcraft II were left for dust as the algorithm reached the dizzy heights of Grandmaster status.
Starcraft II is one of the world's most popular esports, a complex game that is trickier than chess. Players control different alien races to build up their forces and destroy opponents, and there are thousands of possible moves at any moment.
The victorious AI algorithm was ranked among the top 0.2 per cent of all human players earning lots of brownie points and bragging rights.
It was able to pip some human performers to the top post by being able to learn through the experience of several competing deep neural networks. It refined strategies and counter-strategies and incorporated mind games such as ambushes.
AlphaStar was developed by leading AI firm DeepMind who describe this AI milestone in the peer-reviewed journal Nature.
Couch Potato?
You may well be wondering what the point is of AI playing so many video games. According to DeepMind, it could help develop other AI tool for the benefit of humanity.
"One of the key things we're really excited about is that Starcraft raises a lot of challenges that you actually see in real-world problems," Dave Silver, who leads the lab's reinforcement learning research group, told the BBC.
"We see Starcraft as a benchmark domain to understand the science of AI and advance in our quest to build better AI systems."
Add your Comment
[LOGIN FIRST] if you're already a member.fields are required.