According to Gizmodo:

The program was developed by the Department of Philosophy, Linguistics, and Theory of Science at the University of Gothenburg in Göteborg, Sweden. Its intelligence score is based off of results from standard non-verbal test questions, which are designed to eliminate cultural and linguistic biases by testing knowledge rather than reasoning[*]

*I think this must be a typo and should actually be the other way round.


From the university itself:

2/9: Claes Strannegård. Cognitive computing. I argue that there is a need for a biologically grounded cognitive model that unifies inductive and deductive reasoning and integrates a number of cognitive functions (such as perception, concept formation and learning) that are intrinsically linked to reasoning.


IQ tests are based on two types of problems: progressive matrices, which test the ability to see patterns in pictures, and number sequences, which test the ability to see patterns in numbers. The most common math computer programmes score below 100 on IQ tests with number sequences. For Claes Strannegård, researcher at the Department of Philosophy, Linguistics and Theory of Science, this was a reason to try to design ‘smarter’ computer programmes. ‘We’re trying to make programmes that can discover the same types of patterns that humans can see,’ he says.

asked 17 Feb '12, 02:36

The_Cthulhu_Kid's gravatar image

The_Cthulhu_Kid
2.4k530

edited 17 Feb '12, 05:14


If the agent was designed to pass IQ tests, then this isn't particularly remarkable.

And shouldn't a test that eliminates cultural and linguistic biases test reasoning, not knowledge? Knowledge is culture dependent, reasoning is universal.

link

answered 17 Feb '12, 03:13

jholyhead's gravatar image

jholyhead
4.3k1113

I was thinking the same thing. Possibly a typo?
(17 Feb '12, 03:15) The_Cthulhu_Kid The_Cthulhu_Kid's gravatar image
Actually it would have to be a typo, as pattern recognition is based on reasoning and not knowledge.
(17 Feb '12, 03:30) The_Cthulhu_Kid The_Cthulhu_Kid's gravatar image
I agree that it's probably a typo. But score in IQ tests is strongly based on knowledge about IQ tests, and I haven't seen any test that eliminates this factor.
(17 Feb '12, 17:09) lacucharita ♦ lacucharita's gravatar image

The headline is a bit misleading. The AI scores 150 on number sequences, but 100 on progressive matrices. Here is more detail on a Raven's Progressive Matrices solver (I think this is the one in the article): http://www.sais.se/mthprize/2011/cirillo_strom.pdf
And an article about "Anthropomorphic Artificial Intelligence" by the quoted researcher: http://www.phil.gu.se/posters/festskrift2/mnemo_strannegard.pdf
I need to spend some time reading these. It looks like interesting material.

edit: If you are interested in the RPM test this looks like a good read:
http://kryten.mm.rpi.edu/COURSES/LOGAIS02/carpenter.pdf

edit 2: I finally got around to reading "Anthropomorphic Artificial Intelligence" and found it had a really interesting perspective on how knowledge about the limitations of human reasoning (e.g. working capacity) might be used to improve AIs. Recommended.

link

answered 17 Feb '12, 22:49

rseiter's gravatar image

rseiter ♦
6.6k628

edited 19 Feb '12, 23:31

Gawker does tend toward the sensationalised headlines, but it tends to grab attention ;)
(18 Feb '12, 00:26) The_Cthulhu_Kid The_Cthulhu_Kid's gravatar image
Your answer
toggle preview

Follow this Question via Email

Once you sign in you will be able to subscribe for any updates here

Q&A Editor Basics

  • to upload an image into your question or answer hit
  • to create bulleted or numbered lists hit or
  • to add a title or header hit
  • to section your text hit
  • to make a link clickable, surround it with <a> and </a> (for example, <a>www.google.com</a>)
  • basic HTML tags are also supported (for those who know a bit of HTML)
  • To insert an EQUATION you can use LaTeX. (backslash \ has to be escaped, so in your LaTeX code you have to replace \ with \\). You can see more examples and info here