Thursday, July 21, 2011

We already have good broad theories of intelligence

In Taking "Singularity" Apart Stuart Staniford writes:
"The idea that intelligence will accelerate in future does not follow from the possibility of developing human-level machine intelligence. It assumes a completely unproven 'law of intelligence' that the smarter you are, the easier it is to produce an intelligence even greater. Maybe it works the other way altogether - the smarter you are, the more complex and difficult it is to produce an even greater intelligence. Perhaps there's some fundamental limit to how intelligent it's possible for any agent to be. We have no clue. We haven't so far seen even a single generation of intelligences (us) producing a more intelligent entity, so the whole intelligence explosion idea consists of extrapolating from less than one data point. It's utter speculation"
I'd like to make a distinction between the hypothetical 'law of intelligence' Stuart states and the idea that intelligence is lawful, that there are patterns between the makeup of an entity and how good it will be at some particular mental task. In a simplified case, you could put certain properties of an entity's brain into a set of equations and get back scores related to its learning, focus, creativity, and such.

It is this kind of 'Laws of Intelligence', laws as scientific theory, which would lead to higher intelligences being better able to apply those laws. I think there is a vast amount of evidence that intelligence is lawful in exactly this way.

One law of intelligence is that from certain measures you can score higher by using more raw speed or memory. You might not paint a better painting or understand a deeper metaphor, but with a proper architecture you can certainly make more paintings, consider more metaphors, and learn faster. Even with no further understanding of it's own intelligence, if a digital being could copy itself to another system, it could do twice the work.

This alone pushes the questions of lawfulness and improvement via intelligence into the realm of manufacturing. It seems clear that 100 engineers would be better than 1 at designing systems to acquire resources, process them, and finally process the designers themselves faster.

Going back to raw algorithms, the degree to which higher intelligence helps design still higher intelligence is likely to be domain specific. Designing a version of yourself that is better at genetic engineering is not that different than writing an automatic tool that solves certain molecular problems quickly, in that both rely on the lawfulness of the problem. It will probably be much easier to improve an AI's ability to program than to improve it's ability to paint, as coding is very lawful and painting less so.

The best evidence that greater intelligence leads to greater ability to create intelligence is the basis for Stuarts own belief that "while I cannot say it's a certainty, it appears to be more likely than not that machines will gradually approach human levels of intelligence in the present century." It took billions of years for mindless natural selection to produce human level intelligence and perhaps a few thousand for us. Today already it appears that a thinking thing can look a program and find ways to make that smarter, or find whole news ways to solve the same problem. More specifically, it also appears that those with higher scores in certain metrics are better at it than others.

The trends we see and projections of improved future AI's only make sense if intelligence is something that can be figured out and constructed. A predictable explosion of figuring-out is just a natural consequence of this.

Monday, March 9, 2009

Thank You Mr. Brown

Finally, after three years of fits and starts, I have finished my Science Fiction epic, Xenobiology.

I started Xenobiology because I was inspired by the wonderful blognovel Simon of Space. Back when SoS started, it was an episodic story posted to blogger every couple of days, so that's the example I followed. I intended my story to be purely episodic, like the old star trek shows, where every few episodes the Professor would visit a new planet and describe the life forms there, all against a backdrop of a technologically stifled galactic civilization.
I wrote a post a day for the first 30 days. Then I let myself have a break, which I probably never should have done. I never recovered that momentum.
But I received enough compliments to make me think I could actually write, and over the course of getting the Professor to where I wanted him I had laid the foundation for an actual plot. Like with an ending and everything.

So I joined a local writer's group, failed at NaNoWriMo a few times, and ever so slowly developed my story and missed my self imposed deadlines.
And now, with several months of increased discipline, It is done. Every future post is loaded into blogger, and I'm scheduling them for a post a day as I do final corrections.

Of course, there is more to the story, but for now I just wanted to thank the author that really got me started in writing.

Friday, February 20, 2009

Soon I meet The Westboro Baptists

The Westboro Baptist Is coming to Buffalo! "On Sunday at 1:45, they’ll be at St. Joseph’s Church at 3275 Main Street in Buffalo near UB South campus."

Update: Photos - The counter-protest by a variety of Buffalo groups was overwhelming. Only 3 of the Westboro Baptists showed up, and they left after a few minutes of being completely surrounded.

Sunday, February 24, 2008

Who Mourns for Corn?

A giant contraption
moves with a dull monotone
incapable of action
they die by the millions; alone.

It is not a sad scene
that they die just now,
if not killed by machine
they'd fall prey to a cow.

They are not complete
or happy at all
until their seeds we eat
picked in a time we call fall.

Perfect contentment
they do enjoy
so do not lament
any plant but for soy.

Friday, October 5, 2007

Justice vs. strict constructionism

My response to this:
The constitution is not an end, it is a means. As a means, it has several ends. One of its stated ends is to ‘establish justice’. Rights are not granted by the constitution, but some are recognized by it explicitly.
If the constitution fails to bring justice, or to recognize a right, the courts still have a responsibility to be just, because justice is the goal, not legalism.

Saturday, September 1, 2007

Do not Question!!!

Measured Against Reality has a post on meaningless questions, questions that are unanswerable because they are poorly formed. Like 'What is outside of the universe?'.

He uses a lot of examples from the debate over evolution and the big bang. When people disagree on a complex subject, they often challenge the other side with questions. They find something that the theory doesn't seem to explain. Sometimes it's a good challenge, sometimes it has an answer, and sometimes it's gibberish.

The problem is that such questions don't have a place in rational discourse, at all.

When you make an argument, you start with premises, follow a logical pattern, and reach a conclusion. If your premises are true and your logic valid, then your conclusion must be true.
For someone to attack your claim, they should make an argument that one or more of your premises are false, or that your logic is invalid. Their argument should also have premises and logic, and be subject to the same style of attack.

Challenge questions sidestep all that. They attack a conclusion without attacking the premises or logic that lead up to it. They make an argument, but make it difficult to criticize by obscuring its implied premises, which may be gibberish. It is this easy and hard to counter attack that makes them popular rhetoric, but bad argument.

Questions for clarification or knowledge are fine. I am only against questions used to attack structured arguments.

We should not ask 'Well then who did Cain marry?'. It provides no premises, and disproves nothing. We should make assertions about who would have been available for Cain, and those assertions can be rationally discussed.

Conversely, we should not answer challenge questions, whether they are meaningful or not. We should put it upon the challenger to make a good argument against our premises or logic. We have no obligation to be sniped at, and if our arguments are good, then theirs will fall flat.

Wednesday, August 22, 2007

A Thwarted Creation Myth

"Rolos are one of the seven perfect candies God created at the beginning of the world."

"That's stupid, I remember when Rolos came out."