Sound and Efficient Closed-World Reasoning for Planning

Abstract Closed-world inference—an essential component of many planning algorithms—is the process of determining that a logical sentence is false based on its absence from a knowledge base, or the inability to derive it. We describe a novel method for closed-world inference and update over the first-order theories of action used by planning algorithms such as nonlin, tweak and ucpop . We show the method to be sound and efficient, but incomplete. In our experiments, closed-world inference consistently averaged about 2 milliseconds while updates averaged approximately 1.2 milliseconds. Furthermore, we demonstrate that incompleteness is nonproblematic in practice, since our mechanism makes over 99% of the desired inferences. We incorporated our method into the xii planner, which supports our Internet Softbot (software robot). The technique cut the number of actions executed by the Softbot by a factor of one hundred, and resulted in a corresponding speedup to xii .

[1]  Hector J. Levesque,et al.  All I Know: A Study in Autoepistemic Logic , 1990, Artif. Intell..

[2]  Christos H. Papadimitriou,et al.  Games against nature , 1985, 24th Annual Symposium on Foundations of Computer Science (sfcs 1983).

[3]  Maria Gini,et al.  An empirical study of sensing and defaulting in planning , 1992 .

[4]  David Chapman,et al.  Planning for Conjunctive Goals , 1987, Artif. Intell..

[5]  Amihai Motro,et al.  Integrity = validity + completeness , 1989, TODS.

[6]  Peter F. Patel-Schneider,et al.  "Reducing" CLASSIC to Practice: Knowledge Representation Theory Meets Reality , 1999, Artif. Intell..

[7]  Marianne Winslett,et al.  Updating logical databases , 1990, Cambridge tracts in theoretical computer science.

[8]  Nils J. Nilsson,et al.  Artificial Intelligence , 1974, IFIP Congress.

[9]  Oren Etzioni,et al.  Building Softbots for UNIX (Preliminary Report) , 1992 .

[10]  Robert C. Moore A Formal Theory of Knowledge and Action , 1984 .

[11]  Sam Steel,et al.  Integrating Planning, Execution and Monitoring , 1988, AAAI.

[12]  Oren Etzioni,et al.  Statistical methods for analyzing speedup learning experiments , 2004, Machine Learning.

[13]  Oren Etzioni,et al.  Intelligence without Robots: A Reply to Brooks , 1993, AI Mag..

[14]  Hirofumi Katsuno,et al.  On the Difference between Updating a Knowledge Base and Revising It , 1991, KR.

[15]  Oren Etzioni,et al.  An Approach to Planning with Incomplete Information , 1992, KR.

[16]  E. Pednault Toward a mathematical theory of plan synthesis , 1987 .

[17]  Matthew L. Ginsberg,et al.  Readings in Nonmonotonic Reasoning , 1987, AAAI 1987.

[18]  Arthur M. Keller,et al.  On the Use of an Extended Relational Model to Handle Changing Incomplete Information , 1985, IEEE Transactions on Software Engineering.

[19]  Daniel S. Weld An Introduction to Least Commitment Planning , 1994, AI Mag..

[20]  Illah R. Nourbakhsh,et al.  Time-Saving Tips for Problem Solving with Incomplete Information , 1993, AAAI.

[21]  Alon Y. Halevy,et al.  Queries Independent of Updates , 1993, VLDB.

[22]  John McCarthy,et al.  Circumscription - A Form of Non-Monotonic Reasoning , 1980, Artif. Intell..

[23]  Mark A. Peot,et al.  Conditional nonlinear planning , 1992 .

[24]  Edwin P. D. Pednault,et al.  Synthesizing plans that contain actions with context‐dependent effects 1 , 1988, Comput. Intell..

[25]  Oren Etzioni,et al.  A softbot-based interface to the Internet , 1994, CACM.

[26]  Gabriel M. Kuper,et al.  Updating Logical Databases , 1986, Adv. Comput. Res..

[27]  Johan de Kleer,et al.  A Minimality Maintenance System , 1992, KR.

[28]  Alvaro del Val Computing Knowledge Base Updates , 1992, KR.

[29]  David E. Smith Finding All of the Solutions to a Problem , 1983, AAAI.

[30]  Oren Etzioni,et al.  Planning with Incomplete Information in the UNIX Domain , 1993 .

[31]  Kurt Konolige Circumscriptive Ignorance , 1982, AAAI.

[32]  Georg Gottlob,et al.  On the complexity of propositional knowledge base revision, updates, and counterfactuals , 1992, Artif. Intell..

[33]  Marco Schaerf,et al.  A Survey of Complexity Results for Nonmonotonic Logics , 1993, J. Log. Program..

[34]  Sarit Kraus,et al.  Nonmonotonicity and the Scope of Reasoning: Preliminary Report , 1990, AAAI.

[35]  Austin Tate,et al.  Generating Project Networks , 1977, IJCAI.

[36]  Charles Elkan A decision procedure for conjunctive query disjointness , 1989, PODS '89.

[37]  Jack Minker,et al.  Logic and Data Bases , 1978, Springer US.

[38]  Oren Etzioni,et al.  Tractable Closed World Reasoning with Updates , 1994, KR.

[39]  Vladimir Lifschitz,et al.  Closed-World Databases and Circumscription , 1987, Artif. Intell..

[40]  Matthew L. Ginsberg A Circumscriptive Theorem Prover , 1989, Artif. Intell..

[41]  Oren Etzioni,et al.  Omnipotence without Omniscience: Sensor Management in Planning , 1994, AAAI 1994.

[42]  Edwin P. D. Pednault,et al.  ADL: Exploring the Middle Ground Between STRIPS and the Situation Calculus , 1989, KR.

[43]  Marianne Winslett,et al.  Reasoning about Action Using a Possible Models Approach , 1988, AAAI.

[44]  Yoav Shoham,et al.  Deriving properties of belief update from theories of action , 1992, IJCAI.

[45]  Robert C. Moore,et al.  Formal Theories of the Commonsense World , 1985 .

[46]  Daniel S. Weld,et al.  UCPOP: A Sound, Complete, Partial Order Planner for ADL , 1992, KR.

[47]  Raymond Reiter,et al.  Circumscription Implies Predicate Completion (Sometimes) , 1982, AAAI.

[48]  Gösta Grahne,et al.  The Problem of Incomplete Information in Relational Databases , 1991, Lecture Notes in Computer Science.

[49]  David E. Smith,et al.  Reasoning About Action I: A Possible Worlds Approach , 1987, Artif. Intell..

[50]  Michael A. Taitslin,et al.  A Logic for Data Description , 1989, Logic at Botik.

[51]  Charles Elkan,et al.  Independence of logic database queries and update , 1990, PODS '90.

[52]  Oren Etzioni,et al.  Planning with Execution and Incomplete Information , 1996 .

[53]  Maria Gini,et al.  Deferred Planning and Sensor Use , 1990 .