Learning in a Subsumption Architecture

Learning in a Subsumption Architecture

Like adding new knowledge, because the basic subsumption architecture has no explicit representation of world knowledge, it cannot learn symbolic knowledge. The modification Mataric describes allows the agent to build a map of its spatial world (which in one sense is a form of rudimentary learning) and use that map in planning but all the behaviors associated with using the map (such as recognizing and recording landmarks) had to be coded by a programmer.


Return to the Issues Index for this Architecture.

Return to the top of this architecture.

Go to a Discussion of this issue for multiple architectures.


Current Location: Subsumption-Issues-Learning

Go to NEXT page.