Relevance Based Planning: Why Its a Core Process for AGI

DOI

10.2991/agi.2009.40How to use a DOI?

Abstract

Relevance Based Planning (RBP) is a general method that plans in interaction with a domain simulation and domain specialized procedures. I argue that exploitation of the properties of causality and Euclidean topology which hold in many domains is a critical inductive bias necessary if an AGI (or any intelligent program) is to generalize to new problems and new domains, and is critical to human thought, and that RBP achieves its efficiency by exploiting these properties in a novel and powerful way, faithful to introspection in an example task. RBP is proposed to be implemented as a scaffold within an AGI or a CAD tool for producing intelligent programs, that is to say as a general library procedure that takes as inputs domain specialized procedures such as a domain simulation and procedures for recognizing and dealing with problems within the simulation. Supplied with such procedures as inputs, the RBP scaffold provides a framework that orchestrates plan formation and refinement in such a way that only causally relevant configurations are considered.

Copyright

© 2009, the Authors. Published by Atlantis Press.

Open Access

This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).

Download article (PDF)

TY  - CONF
AU  - Eric B. Baum
PY  - 2009/06
DA  - 2009/06
TI  - Relevance Based Planning: Why Its a Core Process for AGI
BT  - Proceedings of the 2nd Conference on Artificial General Intelligence (2009)
PB  - Atlantis Press
SP  - 190
EP  - 191
SN  - 1951-6851
UR  - https://doi.org/10.2991/agi.2009.40
DO  - 10.2991/agi.2009.40
ID  - Baum2009/06
ER  -