A universal measure of intelligence

Any new measure of intelligence is really just a new definition of intelligence.

Basing an intelligence metric on Kolmogorov complexity seems like a really interesting idea. But Kolmogorov complexity is not computable so I don't understand how it can be used in practice. Anyone know of any free papers that provide insight into this ?

Oh joy, another dubious article from physorg.com!

With the science sub-reddit's recent acquisition of real scientist moderators, I think we have finally reached the point where the science submissions on HN are less credible then those on reddit.

Article is really light on details.. But they mention that they are basing their intelligence metric on an organism's Kolmogorov complexity. It sounds to me like they are quantifying randomness, not intelligence.

I'm guessing they mean something like this: repeatedly pose a random reinforcement-learning problem and record the agent's score. The "IQ" is the Kolmogorov complexity of the problems at which the agent goes from usually acing it to usually bombing. If you pick a random program (as your random problem), its K-complexity is probably near the length of the program, so you don't need to be able to compute the (uncomputable) K-complexity to get a good estimate of the above score.

(Based on reading a few words before about Shane Legg's work in the same area, not yet published AFAIK. This is just a guess since there were very few details on that too.)

Update: from the paper (thanks to rubidium), on first skim, the above is basically right. They sum up weighted rewards instead of looking for a K-complexity that separates good from bad performance -- which I'd thought of but wasn't sure how to weight scores from problems of different complexity.

There can't be an algorithmic test for intelligence, other than by an equally or greater intelligent agent, because anything automated has answers static enough that you can just code specific solutions for them.

I hate when articles like this talk about the research, but don't offer examples of what kind of questions they would ask. I'm much more interested in knowing what kinds of questions they would ask people and how I would do in said questionnaire.