Learning by Analogy in Soar
Steir and Adelson
describe a Soar system that built analogies in
learning how a "stack" computer data type could be learned. This system
was called Data-Soar. A
problem space called Use-Analogy
was established that allowed the comparison of different functionalities
within the space. As relationships were discovered,
chunks were created that captured
the discovered analogy.
This is a simple example of several small tasks that have been
accomplished to demonstrate the capability of analogical learning
in Soar. However, extensive use of analogical learning has not
been investigated.
To go to the Soar capabilities list, press UP
To return, press HOME.
To go to the next document, press NEXT.