Journal of Dw

AI, comparison stage

Tue, 15 Jan 2013 22:46:03 GMT

The "Great Concept" investigation cycle (of either my creatures, or a particular CTC) goes like this:

- Imagine a concept that you want to investigate.
- Build an (imperfect) materialization of this concept.
- Compare the materialization to the concept to learn more about both.
- Repeat.

Now I'm going to do something slightly different, but reminiscent of it, with regards to the general optimizer I wrote about.

It's not a Great Optimizer, because there can't be any single Great Optimizer, and I'm not comparing it to the concept of intelligence, but to ourselves. Still.
( Read more... )
So that's a rather good instance of the investigation cycle, I think. By contrasting, I know more about both the AI and ourselves.

There's an even more subtle, and more general, thing to be found by looking at the optimizer-in-two-realms observation, but I'm not quite awake enough that I can state it here yet. I'm not sure I can see it in its entirety yet, either.

-

[1] Even if it were self-aware in that sense, we have no way of knowing that the self-awareness maps to what it's really doing. If the optimizer is moving a robot body out a door, we have no way of knowing that the animist connection would actually perceive that as moving. What we would call "moving" entails a lot of other changes, e.g. of states in the computer's memory, and perhaps the crossing awareness would be focused on that instead.

comments

AI and consciousness, second half

Thu, 10 Jan 2013 15:01:20 GMT

About consciousness and computers, I gave three possibilities for the effect of consciousness:

- Consciousness might let the conscious being do things no computer can do,
- Consciousness might be one way of doing somthing an ordinary computer could also do,
- or consciousness might not have any effect on the material self at all.

Here I'll deal with the first possibility, which is both good and bad for general optimizers and artificial intelligences.
( Read more... )

comments

AI and consciousness, first half

Wed, 09 Jan 2013 22:24:47 GMT

And now for the second part of what I thought about... here it gets rather less concrete.
( Read more... )

[1] (In the original text file that I wrote, I forgot about another possibility I've later come to call the "CQ" possibility. In brief, perhaps consciousness only recently evolved. If so, the imperfect spread of it, due to evolution not having caught up fully yet, would permit kinda-zombies: less effective but still very real. But this has other problems. Perhaps I'll write more about it in a post of its own after all this :) )

(And is anyone actually reading this? I'm curious.)

comments

AI, part three!

Fri, 04 Jan 2013 15:32:02 GMT

The impractical optimizer layout I gave has been discovered before -- that's where I took it from, actually. It's called AIXI. The predictor system is called an "universal prior" (though it does depend on the assumption that the universe is computable - i.e. the Church-Turing thesis - to be truly universal). The attempt at a more practical system is called MC-AIXI (MC for Monte Carlo). I was dimly aware of it (I even linked to it earlier), but I hadn't actually read the paper.
( Read more... )

comments

AI, part two!

Wed, 02 Jan 2013 22:19:26 GMT

Alright, so that's a completely impractical General Optimizer. How may one make it more effective?
( Read more... )

comments

AI, or Postings from a Train, part 1

Tue, 01 Jan 2013 22:30:05 GMT

In the domain of artificial intelligence, I've been thinking about what one may call a "General Optimizer" (or perhaps Great Optimizer, though there can be no one single Great - in the Platonic/Vasai sense - Optimizer). The General Optimizer is a system that you provide with a fitness function, and then it acts to maximize the value of that fitness function, given the manipulators and actions available to it. What makes the optimizer General is that the fitness function can be anything you can evaluate. It might be the proportion of red pixels in a picture the Optimizer gets from a camera. It might be the fraction of satisfied clauses in a SAT instance; or it might be the average return on some stock market data according to a strategy it devises.
( Read more... )

comments