Ben said:
I think the PLN / indefinite probabilities approach is a complete and
coherent solution to the problem. It is complex, true, but these are
not simple issues...
-------------------------
I just started reading Ben's paper "Indefinite Probabilities for General
Intelligence" and while I congratulate Ben and his colleagues for recognizing
that there is a problem (that is the first step son) I am really skeptical
about the idea that a probability method could be instrumental in solving the
problem.
But first of all: Thank you for providing some examples! If only the other
edg-cated perfessers who wrote pdf papers could attain that level of insight
about how learning actually works. Who knows what might happen. But how can we
commoners expect the noble academics to appreciate the idea that providing
examples for motivated students might actually work to help them learn. That
insight must still be on the frontiers of insight about learning, or maybe that
is just one more area of intelligence that I just don't get, because the vast
majority of the of the puffed up geniuses who write online paper do not provide
examples.
I do not yet understand what was written in the paper, (I will study it since
it has some worked examples) but I think I got the basic direction. Or at
least I can find a basic problem. The problem is is that you will need to
utilize some evaluation of the probability of a distribution model for the
uncertainty of the uncertainties and while that is very very cool, in the real
world you cannot pay the bills with play money (unless you are an unusually
powerful child of a ruler or child-ruler of the world or something along those
lines).
We all have to deal with uncertainty in our own ways. But, the use of false
values of precision is just not going to be an explain-it-all. False levels of
precision would undoubtedly be a major benefit in getting gummint funding, but
the frame of your method of attacking the problem indicates to me that you are
confusing the efficacy of fuzzy methods to produce good results in some cases
with the efficacy of methods that would be needed to produce good results in a
greater variety of situations. So while I am sure that you are an exceptional
teacher, I am also able to assign a made up probability of .96532 that you have
not yet found the yellow brick road.
Jim Bromer
-------------------------------------------
agi
Archives: http://www.listbox.com/member/archive/303/=now
RSS Feed: http://www.listbox.com/member/archive/rss/303/
Modify Your Subscription:
http://www.listbox.com/member/?member_id=8660244&id_secret=103754539-40ed26
Powered by Listbox: http://www.listbox.com
