On 8/7/07 8:49 PM, "John Black" <JohnBlack@xxxxxxxxxxx> wrote: (01)
> How about on page 68, when he says, "The answer is the brain doesn't
> "compute" the answers to problems; it retrieves the answers from memory.
> ....The entire cortex is a memory system. It isn't a computer at all.....Let
> me show, through an example, the difference between computing a solution to
> a problem and using memory to solve the same problem. Consider the task of
> catching a ball. Someone throws a ball to you, you see it traveling toward
> you, and in less than a second you snatch it out of the air. This doesn't
> seem too difficult - until you try to program a robot arm to do the same. As
> many a graduate student has found out the hard way, it seems nearly
> impossible......And although a computer might be programmed to successfully
> solve this problem, the one hundred step rule tells us that a brain solves
> it in a different way. It uses memory......The memory of how to catch a ball
> was not programmed into your brain; it was learned over years of repetitive
> practice, and it is solved, not calculated, in your neurons."
>
DN: Here I would argue that both patterns are in fact computational in
nature. The difference is the architecture that could be used to implement
the physical structure of the system. One is actively computing on the heap
while the other is akin to retrieving the state of something from a
persistence tier. In human terms this might be differentiated in short term
vs. long term memory. In the IT realm, it is the difference between active
memory (stack/heap) and archiving (hard drive). Perhaps the two are not so
different. (02)
Applicaton in computational Intelligence? (03)
What would be useful for CI/AI research is to merely document the high level
design patterns for problem solving. This has been done extensively for
patterns such as the Blackboard AI pattern, Forward chaining etc but I feel
there are some gaps. (04)
At Java one, a student group from Quebec submitted a robotic submarine with
a complex series of algorithms to self level while following a main
directive (a pre-programmed series of objectives). The submarine failed to
self level as the algorithms did not learn and adapt. An example was if the
sub tiled aft at 10 degrees, it was programmed to use thrust in certain
vectors until the level read neutral (zero degrees). The failure on the
programmers part was to realize that by continuing to thrust until a (level
= true) state had been achieved resulted in over compensation until a state
of bow tilt was achieved due to the momentum built up in the vertical plane.
The velocity always kept the self leveling compensation working past the
point it should have shut down. If the algorithm learned and adapted (as
human children do) it could have varied the thrust compensation until it
found a balance where it could achieve the desired state. For this reason,
I gave the 'Duane Award" to the automated helicopter team who had the
foresight to create a learning/adaptive algorithm. (05)
http://technoracle.blogspot.com/search?q=helicopter (06)
This is but one example of how intelligence can be bestowed upon
applications (in my own definitions anyways). (07)
Then again, I am merely a hack at this Ontology stuff. In such esteemed
company, I am sometimes reluctant to write posts out of fear they are way
off base. (08)
Cheers (09)
/d (010)
_________________________________________________________________
Message Archives: http://ontolog.cim3.net/forum/ontolog-forum/
Subscribe/Config: http://ontolog.cim3.net/mailman/listinfo/ontolog-forum/
Unsubscribe: mailto:ontolog-forum-leave@xxxxxxxxxxxxxxxx
Shared Files: http://ontolog.cim3.net/file/
Community Wiki: http://ontolog.cim3.net/wiki/
To Post: mailto:ontolog-forum@xxxxxxxxxxxxxxxx (011)
|