‘Bull’s eye’ algorithm is about to revolutionise computer modelling

18 Nov 201563 Shares

Share on FacebookTweet about this on TwitterShare on LinkedInShare on Google+Pin on PinterestShare on RedditEmail this to someone

Share on FacebookTweet about this on TwitterShare on LinkedInShare on Google+Pin on PinterestShare on RedditEmail this to someone

In what could totally change the very fabric of computer modelling calculations, a team from MIT in the US has created a new “bull’s eye” algorithm that can make modelling 200-times faster than before.

Computer modelling is used in almost every aspect of science and engineering when it comes to predictions or design, by allowing computers to create likely outcomes for, say, monitoring the Earth’s climate or analysing the possibilities of running an advanced engine at a particular setting, according to MIT News.

But, until now, scientists have been largely using a technique when formulating computer models known as the Markov chain Monte Carlo (MCMC) analysis.

This method is often referred to as the ‘Monopoly method’ after the famous board game of the same name. As a player rolls the dice, the more they play, the greater the chance that they are likely to know where they are going to land based off the number they roll.

The same goes for traditional MCMC analysis, which determines a model by running it over and over again and recognising a pattern.

200-times faster

However, the ‘shrinking bull’s eye’ analysis – referred to as such due to its appearance during the modelling process – will continually narrow in on its target, offering a probability distribution of values for each unknown parameter.

The researchers working on the algorithm ran the same test on both the traditional MCMC analysis and its own shrinking bull’s eye method, and with the latter was able to reach the same answer 200-times faster than with the former.

“What this means, in the long run, is things that you thought were not tractable can now become doable,” one of its creators, Youseef Marzouk, said. “For an intractable problem, if you had two months and a huge computer, you could get some answer, but you would not necessarily know how accurate it was.

“Now for the first time, we can say that if you run our algorithm, you can guarantee that you’ll find the right answer, and you might be able to do it in a day. Previously that guarantee was absent.”

One of the first uses of the algorithm has been in simulations of sea ice movement in Antarctica and it has shown itself to be 60-times faster than the method scientists had been using so far.

Future uses of the shrinking bull’s eye method could also one day contribute to the development of even more advanced artificial intelligence capable of responding to situations much faster than it is capable of now.

Bull’s eye image via Shutterstock

66

DAYS

4

HOURS

26

MINUTES

Buy your tickets now!

Colm Gorey is a journalist with Siliconrepublic.com

editorial@siliconrepublic.com