Jump to content

Home

Being Isaac Newton: Computer Derives Natural Law from Raw Data


Achilles

Recommended Posts

Link to the full story

 

Excerpt:

The researchers have taught a computer to find regularities in the natural world that become established laws – yet without any prior scientific knowledge on the part of the computer. They have tested their method, or algorithm, on simple mechanical systems and believe it could be applied to more complex systems ranging from biology to cosmology and be useful in analyzing the mountains of data generated by modern experiments that use electronic data collection.

 

Their process begins by taking the derivatives of every variable observed with respect to every other – a mathematical way of measuring how one quantity changes as another changes. Then the computer creates equations at random using various constants and variables from the data. It tests these against the known derivatives, keeps the equations that come closest to predicting correctly, modifies them at random and tests again, repeating until it literally evolves a set of equations that accurately describe the behavior of the real system.

 

The researchers tested the method with apparatus used in freshman physics courses: a spring-loaded linear oscillator, a single pendulum and a double pendulum. Given data on position and velocity over time, the computer found energy laws, and for the pendulum, the law of conservation of momentum. Given acceleration, it produced Newton’s second law of motion.

 

The researchers point out that the computer evolves these laws without any prior knowledge of physics, kinematics or geometry. But evolution takes time. On a parallel computer with 32 processors, simple linear motion could be analyzed in a few minutes, but the complex double pendulum required 30 to 40 hours of computation. The researchers found that seeding the complex pendulum problem with terms from equations for the simple pendulum cut processing time to seven or eight hours. This “bootstrapping,” they said, is similar to the way human scientists build on previous work.

 

Computers will not make scientists obsolete, the researchers conclude. Rather, they said, the computer can take over the grunt work, helping scientists focus quickly on the interesting phenomena and interpret their meaning.

 

Pretty neat stuff. It would be awesome if someday computers can be used to crack huge data sets that baffle human scientists. Of course, it would be a little weird to see a supercomputer get the Nobel Peace Prize for its ground-breaking work on string theory :p

 

Of course, we'll all be running from Skynet whilst plugged into The Matrix by then anyway, so...

Link to comment
Share on other sites

This may or may not be off-topic, but how close do you think we are to true artificial intelligence? Granted that the computer necessary for such a thing would have to be the size of the Empire State Building, but still, do you think that we'll live to see it?

Link to comment
Share on other sites

This may or may not be off-topic, but how close do you think we are to true artificial intelligence? Granted that the computer necessary for such a thing would have to be the size of the Empire State Building, but still, do you think that we'll live to see it?
In my complete and utter non-expert opinion? It's impossible to say.

 

In order to understand what to build, we need to understand how our own brains work in the first place. And that endeavor is so close to the starting line that it's difficult to argue that we've even really begun. But the reality is that they're called "breakthroughs" for a reason.

Link to comment
Share on other sites

Whoa. I always like these discussions. Cool thread.

 

Yeah, busy work is always what seemed to frustrate me in school. I'm happy that now the frustrating, tedious busy work is up to machine precision that is not nearly as subject to fatigue errors. Scientists can get to the good stuff. The complex conundrums.

 

[speculative wishing] Would it not be such a grand thing to have some kind of convenient interface giving a sudden boost to one's own intelligent computing? [/HK-47]

 

I can imagine at some point, though, our progress surge is going to hit a margin where it will taper off. Then soon it will hit a dead stop...assuming other things have not failed first. But not to spoil the moment, let us ride the surge sky high and enjoy it!!!

 

Still, I can just imagine what this means. A simple visor computer, for say, a welder. You could see in real time all sorts of little things how to go about making those tricky welds, compute the stresses it is likely to encounter...and a bazillion other things I could mention but won't for it would detract too much from the subject at hand.

 

This is just, so amazing I can't even begin to think of all the wonderful things, uses and interfaces that this breakthrough could possibly mean. Not to mention the vastly tremendous increases in performance and work quality this could lead to. Seriously I am already abuzz with little devices that probably will be made now with on board computer systems. Kind of like all those bands and visors in the KOTOR games.

 

A jeweler could analyze shatter points of gems quicker. Safety inspectors no longer needing mechanical/structural engineering degrees to see a structure is a disaster...before it is ever built or harms thousands...

 

This has set my brains on fire with all kinds of thoughts of possible uses. Positively tingling!

Link to comment
Share on other sites

Quite the contrary. Was elaborating. If it is doing the busy work while the scientists figure out the real, complex problems. Well, that would be awesome. We'd figure out so much more, so much quicker...... to a point. Whether we would

1) just come up on some order of magnitude, so large I don't care to try to guess how big, and screech to a stop,

 

OR

 

2) some other thing happens and everything stops (IE natural disaster setting us back several thousand years)

I'm not sure. I do know, research speed is soon to go from 0 to 999,999,999 in 0.02 seconds according to your article.

 

Then I theorized about how such things might be nice as wearable enhancements for us human beings.

Link to comment
Share on other sites

Archived

This topic is now archived and is closed to further replies.

×
×
  • Create New...