Learning heuristics for ordering plan goals through static operator analysis

T. L. McCluskey, J. M. Porteous

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

There is a trade-off between the generality and efficiency of automatic planning systems which means that general planners tend to be inefficient, the problem of “efficiency versus generality”. Here we present PRECEDE, a novel method of off-line compilation, that analyses domain operators and produces heuristics that reduce search during plan generation. We present a declarative specification of PRECEDE, illustrate it with a worked example and discuss the results of some empirical tests and related work.

Original languageEnglish
Title of host publicationMethodologies for Intelligent Systems
Subtitle of host publication8th International Symposium, ISMIS '94 Charlotte, North Carolina, USA, October 16–19, 1994 Proceedings
EditorsZbigniew W. Raś, Maria Zemankova
PublisherSpringer-Verlag Berlin Heidelberg
Pages406-415
Number of pages10
VolumeLNCS 869
ISBN (Electronic)9783540490104
ISBN (Print)9783540584957
DOIs
Publication statusPublished - 1994
Event8th International Symposium on Methodologies for Intelligent Systems - Charlotte, United States
Duration: 16 Oct 199419 Oct 1994
Conference number: 8
https://www.springer.com/gp/book/9783540584957

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume869 LNCS
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Conference

Conference8th International Symposium on Methodologies for Intelligent Systems
Abbreviated titleISMIS 1994
CountryUnited States
CityCharlotte
Period16/10/9419/10/94
Internet address

Fingerprint Dive into the research topics of 'Learning heuristics for ordering plan goals through static operator analysis'. Together they form a unique fingerprint.

Cite this