It’s been a good week for robot scientists. Adam, a robot developed by Aberystwyth University, ran a whole experiment and made a new discovery without human help. It’s the first independent robot discovery.
Now researchers from Cornell claim to have invented a computer that can “derive natural laws from observed data.” Without knowing anything about science, this computer looks at simple mechanical systems, runs an algorithm and figures out the laws that govern it.
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 behaviour of the real system.
Technically, the computer does not output equations, but finds “invariants” — mathematical expressions that remain true all the time, from which human insights can derive equations.
It has independently and without previous knowledge found energy laws and Newton’s second law of motion. When the researchers bootstrap scientific knowledge into the machine, the process takes much less time.
Although it’s currently only looking at simple things, used in first year physics courses, there seems to be no reason why this computer could not analyse complex phenomena. That is, nature itself.
Now that the robots are doing the jobs of scientists, scientists can get back to doing the job of robots. That is, destroying all mankind.
2 comments
Exploding Bat
April 10, 2009 at 1:35 am (UTC -4) Link to this comment
Pedantically, the first robot discovery was the proof of the Four Colour Map Theorem.
Ryan Oakley
April 10, 2009 at 9:08 am (UTC -4) Link to this comment
Got me — changed it to the “first independent robot discovery.”