Computers

Artificial intelligence program that learns like a child

Artificial intelligence progra...
An artificial intelligence program created at the University of Gothenburg imitates a child's cognitive development to learn basic arithmetic, logic, and grammar with no pre-existing knowledge (Image: Shutterstock)
An artificial intelligence program created at the University of Gothenburg imitates a child's cognitive development to learn basic arithmetic, logic, and grammar with no pre-existing knowledge (Image: Shutterstock)
View 1 Image
An artificial intelligence program created at the University of Gothenburg imitates a child's cognitive development to learn basic arithmetic, logic, and grammar with no pre-existing knowledge (Image: Shutterstock)
1/1
An artificial intelligence program created at the University of Gothenburg imitates a child's cognitive development to learn basic arithmetic, logic, and grammar with no pre-existing knowledge (Image: Shutterstock)

Artificial intelligence programs may already be capable of specialized tasks like flying planes, winning Jeopardy, and giving you a hard time in your favorite video games, but even the most advanced offerings are no smarter than a typical four-year-old child when it comes to broader insights and comprehension. It makes sense, then, that researchers at the University of Gothenburg have developed a program that imitates a child's cognitive development.

"We have developed a program that can learn, for example, basic arithmetic, logic, and grammar without any pre-existing knowledge," says Claes Strannegård. Starting from a set of simple and broad definitions meant to provide a cognitive model, this program gradually builds new knowledge based on previous knowledge. From that new knowledge it then draws new conclusions about rules and relations that govern the world, and it identifies new patterns to connect the insight to.

The process is similar to how children develop intelligence. A child can intuit, for example, that if 2 x 0 = 0 and 3 x 0 = 0 then 5 x 0 will also equal 0, or they could draw the conclusion that the next number in the series "2, 5, 8" will be 11. And the same kinds of intuition carry across to other areas, such as grammar, where it's easy to identify rules for standard verb conjugations from examples like sing becoming sang and run becoming ran in the past tense.

"We postulate that children learn everything based on experiences and that they are always looking for general patterns," Strannegård says.

The researchers' system, which they call O*, follows the principle of Occam's razor – that you should favor short and simple explanations over long and complex ones. It identifies patterns by itself and combines them with prior knowledge to solve problems.

Sometimes this will lead to errors, such as when children say "I brang my lunch" instead of "I brought my lunch," but O* managed not only to learn arithmetic from scratch, but also to perform above the average human level on propositional logic problems. And given enough information the researchers hope their program could learn and reason its way to correct conclusions across a range of domains without any need for a programmer to explicitly formulate which rules it should apply in a given situation.

"We are hoping that this type of program will eventually be useful in several different practical applications," says Strannegård. "I think a versatile household robot would be tremendously valuable, but we’re not there yet."

Strannegård and his colleagues presented a paper describing O* at the Seventh Conference on Artificial General Intelligence in August.

Source: University of Gothenburg

5 comments
Nairda
Well written. However even after reading their white paper beyond this article, I could not find references to equipment used to generate these results. Based on the simple approach taken to general intelligence, one may believe it was achieved with minimal hardware over a short period of time. However this assumption may be completely incorrect, and it may have been a supercomputer running these rules for hours on end. A bit more clarity on platform resources and time to achieve these results would have been appreciated. I do appreciate most kids take a little time to learn without distractions and this machine does learn it to a greater depth so not judging either way.
someguy
In 1985, they already had a similar idea in "A Mind Forever Voyaging"...
ezeflyer
AI could be a Terminator technology or our savior. It would depend on substituting non-bribe taking computers for bribe taking politicians.
Larry Hooten
I want one! I wonder how much memory it uses?
Martinjuk
@ Larry - lol, me too! I'm undecided about AI; still not sure whether to rejoice to it or to worry about our future. I came across many articles on this topic, since this is something that interests me a lot (not to mention my boy-ish dream of building one of my own). And it seems that we may be even closer to proper AI than we thought. And should we live long enough to see it become our reality, I just hope we don't end up as exibits in a museum or in a ZOO, but to co-exist in harmony with these way-more-intelligent new species. Food for thought - http://artificial-intelligence.com/comic/14 lol