Thursday, October 25, 2012

Managing expectations in education

Only small amounts of learning are possible since learning is an NP complete problem.

Friday, October 12, 2012

Mind and brain

If a computer is a reconfigurable causal network then perhaps a mind can be best thought of as a self reconfigurable causal network (or a self reconfigurable influence network).

Friday, October 5, 2012

Quantum computers

If a conventional computer is a (large) reconfigurable causal network (see my blog of  28 August       2012) perhaps a quantum computer should be thought of as a reconfigurable influence network.

Monday, October 1, 2012

Value change

It may be that values should change on multiple time scales.

My Asa H architecture is topped by a value module (see Trans. Kansas Acad. Sci., vol. 109,
no. 3/4, page 159, 2006, figure 1).  This can consist of a value network like that in my 21 Sept. 2010 blog. Slow (long term) value change can be made by turning this Bayesian network into a (time varying) dynamic Bayesian network.

A module for (an additional) more rapid value change might be built from a subroutine that measures recent success (utility and utility change), a subroutine that measures current risk tolerance, a subroutine that measures recent degree of change in the knowledgebase, a subroutine that measures current system damage, a subroutine that measures current needs (like energy, hardware, etc.), etc., and a routine that selects (and resolves conflicts between) these subroutines.

Which devices?

Rather than trying to have one device (a smart phone?) do everything I tend to prefer having many devices, each optimized for doing one or two things.

The miscellaneous category (in a casebase)

If a case is "far enough" away from all existing clusters (categories), and  <N other cases have been found that are "near" to it, then add this case to a "miscellaneous" category.

Search and employ the miscellaneous category as if it was a cluster of its own. (But do not define a mean and standard deviations for it. It is not compressible.)

Remove a case from "miscellaneous" and form a new cluster if and when N is exceeded (i.e., when enough cases like this one are discovered).  (A reasonable value for N might be estimated/arrived at by looking at the number of cases found in all the other categories; the mean and standard deviation of this number. N should be set a few standard deviations below the mean. The size of the "miscellaneous" category should also be kept similar to the size of other categories.)