Free Republic
Browse · Search
News/Activism
Topics · Post Article

To: Aric2000
Some can, and some can't, because some will not publish, nor create until a patent is issued.

Okay, but the statement did not limit itself. In any case, the original topics of discussions were two. One involving the cubic function and a granted patent to a human designed circuit. This assertion was made by a poster

Deal with it. In this case, it made a cubic function generator circuit which outperforms the best that all electronic engineers were capable of producing in all the history of electronics.

The circuit at the top was patented in 2000, and is the current state of the art. The circuit at the bottom was produced by pure unaided evolution, and outperforms the human version.

I asked for evidence of this performance edge. None has been produced. I found the patent and examined the circuit and it involves 5 transistors and 4 diodes. The circuit used in the "runoff" has 9 transistors. Somebody changed something. Despite this not one graph comparing the two circuits has been produced here. I suspect that the performance edge is a paper product. Something that the emulating program has produced. Why do I surmise that? Because in evolving the circuit I doubt that each individual circuit was constructed in order to measure its performance of the cubic function. That would be impractical.

From the article ---A genetic programming run typically spawns a population of tens or hundreds of thousands of individuals that evolve over dozens or hundreds of generations. A weeklong run on a laptop computer is sufficient to produce half of the human-competitive results listed in the box on the preceding page; however, all six of the inventions patented after 2000 required more horsepower than that.

The second topic relates to the assertion that many patents have been issued for inventions produced by genetic programming. The same article that had the cubic function circuit in it also made mention of a patent application ---

We have filed a patent application that covers both the new rules and the new controller topologies. If (as we expect) the patent is granted, we believe that it will be the first one granted for an invention created by genetic programming.

This was in the article with the date of February 2003. Clearly, it has not been granted. It does not matter if it is still pending for it to be evidence that there are no patents for inventions created by genetic programming. That still leaves the ability to produce the evidence of many such inventions, but no one has done that.

1,236 posted on 05/11/2003 10:57:35 PM PDT by AndrewC
[ Post Reply | Private Reply | To 1235 | View Replies ]


To: AndrewC
These programs do not face a problem which "real" evolution would, namely extinction of species. It is never an option for the run to go extinct.
1,237 posted on 05/11/2003 11:02:26 PM PDT by HiTech RedNeck
[ Post Reply | Private Reply | To 1236 | View Replies ]

To: AndrewC
You are going to make me go search for those patents aren't you?

Geez, they exist AndrewC, and pretending they don't because I can't get you a patent number RIGHT NOW, is not going to make that fact go away.

Give me a couple of days, I have other things I have to do with my time, and I will do my best to get the relevant info for you.

But the fact of the matter is, even after I give you the patent number and the info that goes with it, you are going to go on a tangent about how it still doesn't prove that evolutionary/genetic programming actually works.

I know you, and your mind is already convinced of the answer, and no matter how many facts are layed out before you, your mind will not be changed.
1,238 posted on 05/11/2003 11:04:54 PM PDT by Aric2000 (Are you on Grampa Dave's team? I am!! $5 a month is all it takes, come join!!!)
[ Post Reply | Private Reply | To 1236 | View Replies ]

Free Republic
Browse · Search
News/Activism
Topics · Post Article


FreeRepublic, LLC, PO BOX 9771, FRESNO, CA 93794
FreeRepublic.com is powered by software copyright 2000-2008 John Robinson