Jump to the Comparative Properties page.
Jump to the Home page.
LEARNING
Learning is the ability of the architecture to acquire domain
specific knowledge that it can later use to improve its performance.
We are specifically concerned with a few different types of learning,
namely:
- Deliberative
- The architecture has a special "learning mode" or needs to be
told what specific things it would be useful to learn.
- Reflexive
- The architecture is capable of learning while doing and freely
intermixes the two. It does not try to learn only when it needs do, it
does so continously.
- Monotonic
- All learning consists of the addition of new facts to the
database. No facts can be deleted.
- Non-Monotonic
- Facts can be deleted from the database.