Much of what we do in perceiving things or thinking is to compress information by finding patterns that match each other and merging or ‘unifying’ them.
Menai Bridge (PRWEB UK) 10 September 2013
The SP theory of intelligence, described in a new article in the journal "Information", aims to solve two problems with computers—their complexity and their lack of intelligence. The theory borrows a neglected idea from psychology: that much of what we do in perceiving things or thinking is to compress information by finding patterns that match each other and merging or ‘unifying’ them.
A simple example is how, if we close our eyes for a moment and open them again, we normally see the same as what we saw before. This is quite different from an old-style cine camera that simply records a succession of frames without any analysis or intelligence. By contrast, we automatically and unconsciously merge successive views into one.
In the SP computer model, based on the SP theory, the matching and unifying of patterns provides the foundation for a concept of "multiple alignment" which is itself the basis for several aspects of intelligence. These include the learning of new knowledge, the processing of language, recognition of patterns, several kinds of reasoning, the retrieval of knowledge, planning tasks, and solving problems.
One relatively simple model does several different things, meaning less complexity than when each aspect is programmed separately. And aspects of the theory—the multiple alignment concept in particular—promise new insights in both psychology and neuroscience.
“Our aim now is to create an "SP machine", a more powerful and robust version of the SP computer model, hosted on an existing high-performance computer,” said Dr. Gerry Wolff, the author of the SP theory. “This would be a facility for researchers everywhere to see what can be done with the system, and to create new and improved versions of it.”
“This research could lead quite soon to improved kinds of database system, with more versatility and intelligence than ordinary databases,” said Dr. Wolff. “Further ahead, there are many possibilities for simpler and more intelligent kinds of computer, and new lines of research in psychology and neuroscience.”
The SP theory is described in: The SP theory of intelligence: an overview (J G Wolff, Information, 4 (3), 283-341, 2013, doi:10.3390/info4030283, bit.ly/1c7KKd1).
More information about the theory may be found on http://www.cognitionresearch.org/sp.htm.
CognitionResearch.org, founded by Dr. Gerry Wolff, conducts research on the SP theory, the SP machine, and related issues in computing and in human perception and thinking. Part of the inspiration for the SP theory has been earlier research by Dr. Wolff developing computer models of language learning and insights gained from work in the software industry.
