First robot capable of adapting!?!

[Zythryn]

I found this fascinating and a bit scary:
http://www.physorg.com/news82910066.html

While this isn't really an A.I. could it be the first step to a robot that can freely adapt and repair itself?

[hallenrm]

Originally Posted by Zythryn

I found this fascinating and a bit scary:
http://www.physorg.com/news82910066.html

While this isn't really an A.I. could it be the first step to a robot that can freely adapt and repair itself?

I really do not find it scary, only interesting news. The claim need to be tested thoroughly, especially the usage of the term Know.

I can indeed imagine that such machines are feasible, but whether it is possible with the present state of programming tools and hardware, is an open question.

[alexander]

whatever hallenrm says

[CraigD]

Originally Posted by hallenrm

I can indeed imagine that such machines are feasible, but whether it is possible with the present state of programming tools and hardware, is an open question.

My read of the linked article give the impression that the described robot mechanicals and programming are complete, making the question closed.

How much, if any, this is a step in the direction of making a computer program with human-like qualities, is, IMHO, the open question of interest here.

[moo]

From the article: "Instead of giving the robot a rigid set of instructions, the researchers let it discover its own nature and work out how to control itself, a process that seems to resemble the way human and animal babies discover and manipulate their bodies. The ability to build this "self-model" is what makes it able to adapt to injury."

Methinks the key phrase here is "seems to resemble" human and animal qualities. While this sounds impressive at first glance, it's doing nothing more than cycling through possibilities until it finds solutions within the parameters of its directive (similar to a "brute force" password cracker, just a bit more complicated).

If results start to fall outside those parameters (due to the "injury" mentioned), it simply starts the process over with modified options (i.e. one less leg).

At best I'd guess it uses a simple neural network, at worst, nothing more than lengthy "switch/case" scenarios (or the equivalent in whatever language is used). I see no evidence of programming magic here, just a novel application.

Don't get me wrong, I do think it's a nice piece of work (with real potential for practical use), but a bit misrepresented as to having any kind of cognizance.

Just my opinion though, perhaps I overlooked something.

moo

[Zythryn]

I don't believe there is any misrepresentation as there is no claim that the robot is self aware or cognizant.

Nor is it programming 'magic'. It is simply programing on a level/direction that has not been done before.

Seems to me it is a first step in the ability to adapt. That ability to adapt is what exites me. I do agree, if it is a simple switch case statement then it is not so impressive. However, if the robot reviews each case and simply tries a select few, or one that works, that is something else.

Basically, that is what animals do when encountering a problem.

[moo]

I don't believe there is any misrepresentation as there is no claim that the robot is self aware or cognizant.

Actually that was exactly the impression I got from the following statement:

In a way, Bongard said, the robot is "conscious" on a primitive level, because it thinks to itself, "What would happen if I do this?"

Far as I can tell, it's no more "conscious" than any other problem solving app, just a bit more sophisticated than some. And IMO it isn't "adapting" any more than your car's onboard computer does when adjusting fuel/air/timing/etc. in response to varying weather and altitude conditions.

moo

[alexander]

robot being able to do a random task and see the results is not any proof of concious behavior, just that there are no onther ways to describe it. The robot does not think "ooh let me go left" the robot goes, random, go left, hit wall, no good, turn around, now go random again. Not quite all the way concious, there is no conginiton that seems to be observed, at least the way it looks to me...