Is There Any Need for Domain-Dependent Control Information?: A Reply
Abstract
In this paper, we consider the role that domaindependent control knowledge plays in problem solving systems. Ginsberg and Geddis (Ginsberg & Geddis 1991) have claimed that domaindependent control information has no place in declarative systems; instead, they say, such information should be derived from declarative facts about the domain plus domain-independent principles. We dispute their conclusion, arguing that it is impractical to generate control knowledge solely on the basis of logical derivations. We propose that simplifying abstractions are crucial for deriving control knowledge, and, as a result, empirical utility evaluation of the resulting rules will frequently be necessary to validate the utility of derived control knowledge. We illustrate our arguments with examples from two implemented systems. Introduction In a AAAI paper entitled "Is there any Need for Domain-Dependent Control Information?" Ginsberg and Geddis (1991) (henceforth G&G) consider whether domain-dependent c...
Cite
Text
Minton. "Is There Any Need for Domain-Dependent Control Information?: A Reply." AAAI Conference on Artificial Intelligence, 1996.Markdown
[Minton. "Is There Any Need for Domain-Dependent Control Information?: A Reply." AAAI Conference on Artificial Intelligence, 1996.](https://mlanthology.org/aaai/1996/minton1996aaai-there/)BibTeX
@inproceedings{minton1996aaai-there,
title = {{Is There Any Need for Domain-Dependent Control Information?: A Reply}},
author = {Minton, Steven},
booktitle = {AAAI Conference on Artificial Intelligence},
year = {1996},
pages = {855-862},
url = {https://mlanthology.org/aaai/1996/minton1996aaai-there/}
}