Communications of the ACM
Computers May Need Patterns to Think Better
Credit: iStockPhoto.com
An international research team, led by Charles Sturt University professor Terry Bossomaier and University of Sydney professor Allen Snyder, is investigating patterns that help computers think better. They say a major difference between human thinking and computational intelligence is that humans use a vast library of patterns, divided up into tens of thousands of smaller chunks of patterns, to develop an intuitive grasp of a situation. The researchers aimed to develop a deeper understanding of how these chunks are gained and how they change with experience.
"We decided to use the ancient oriental game of Go and study how experts in Go use patterns to remember strategies for the game, and how these might be simulated in a computer program," Bossomaier says.
The first phase of the research found that human knowledge undergoes dramatic reorganization moving from amateur to professional ranks. "The change takes place not just in the areas of ‘deep strategy,' where one would expect the big gains to be, but also there is a radical reorganization at the perceptual level," Bossomaier says.
From Charles Sturt University
View Full Article
Abstracts Copyright © 2012 Information Inc., Bethesda, Maryland, USA
No entries found