Black box vs. Glass box knowledge representation:

Black box knowledge representation may be defined as the inability of rules to examine other rules. Architectures with this property are limited in their direct inferencing ability to see what other rules are up to. There are advantages to this:

1. rule bases might be less fragile to changes
2. rule bases might be more modular and easier to understand.

Glass box knowledge representation may be defined as the ability of rules to examine each other. In architectures with this property meta-reasoning (or reasoning about reasoning) may be sped up. Also, the rules and the architecture may share the responsibility to examine, activate and rewards other rules. The advantages of this are:

1. increased flexibility
2. faster meta-reasoning

Black-box knowledge representation does not rule out meta-reasoning, but it would make it more circuituitous. Rules would have to observe each others effects and infer conditions.

 Examples of Black box knowledge representation architectures are:

Atlantis by E. Gat.
Behavior-Based Programming by R. Brooks.
ERE by Drummond et al.
SOAR by A. Newell et al.
Subsumption Architecture by R. Brooks



 Examples of Glass box knowledge representation architectures are:

Dynamic Control Architecture by B. Hayes-Roth.
Homer by Vere & Bickmore.
Icarus by Langley.
MAX by Kuokka et al
Prodigy by Carbonell et al
RALPH by Ogasawara and Russell.
Teton by VanLehn & Ball.
Theo by Mitchell et al




Other Properties.

Back to the Title Page.