New Scientist recognises big “leap forward for artificial intelligence”
Posted: 22nd December 2010
“AS YOU read this article, your brain not only takes in individual words, but also combines them to extract the meaning of each sentence. It is a feat any competent reader takes for granted, but it's beyond even the most sophisticated of today's computer programs. Now their abilities may be about to leap ahead, thanks to a form of graphical mathematics borrowed from quantum mechanics.”
So starts an article in a recent edition of New Scientist magazine, highlighting the work of Oxford University Computing Laboratory in "Quantum Linguistics" by Bob Coecke, Mehrnoosh Sadrzadeh, Ed Grefenstette and Stephen Pulman, drawing from earlier work by Samson Abramsky and Bob Coecke on Quantum Computing.
Featured on the front page of the illustrious publication, the article describes how the department’s Quantum Group along with the Linguistics Group are enabling computers to better understand language by encoding words and grammar in a set of mathematical rules.
At the moment computers only understand sentences as a collection of different words without any structure – a computer would find it very hard to tell the difference between 'Jane likes cheese' and 'Jane does not like cheese.' This pioneering the new approach to linguistics uses a graphical form of category theory – a branch of mathematics that allows different objects within a collection, or category, to be linked – that was developed for use in quantum mechanics. Despite the similarity of words in the above sentences; the new vector representations of these two sentences would show their fundamental difference in meaning.
Bob Coecke likens the situation to watching television at a pixel level. "Rather than seeing the image, you get it in terms of 0s and 1s," he says. "It wouldn't mean anything to you." By translating quantum mechanical processes into pictures, higher-level structures become visible.
This research has many practical applications, for example in the development of search engine technology.
The full article is available on the New Scientist website: http://www.newscientist.com/article/mg20827903.200-quantum-links-let-computers-understand-language.html
The research will be presented at the International Conference on Computational Semantics in Oxford in 2011. http://www.meaningfactory.com/iwcs2011/index.html