A model generation theorem prover in KL1 using a ramified-stack algorithm:

Abstract: "This paper presents a model-generation based parallel theorem prover for first-order logic implemented in KL1. The prover, called MGTP, is efficient in solving range-restricted non-Horn problems. The range-restrictedness condition allows us to represent object-level variables with KL...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Fujita, Hiroshi (VerfasserIn), Hasegawa, Ryuzo (VerfasserIn)
Format: Buch
Sprache:English
Veröffentlicht: Tokyo, Japan 1990
Schriftenreihe:Shin-Sedai-Konpyūta-Gijutsu-Kaihatsu-Kikō <Tōkyō>: ICOT technical report 606
Schlagworte:
Zusammenfassung:Abstract: "This paper presents a model-generation based parallel theorem prover for first-order logic implemented in KL1. The prover, called MGTP, is efficient in solving range-restricted non-Horn problems. The range-restrictedness condition allows us to represent object-level variables with KL1 variables, and to use only matching rather than full unification, thereby obtaining an interpreter that is very simple yet efficient. The implementation techniques developed are also useful for other related areas, such as truth maintenance systems and intelligent database systems. To improve the efficiency of MGTP, we also developed a ramified-stack (RAMS) algorithm for removing redundant computation during conjunctive matching phase of a proving process
Experimental results show that an MGTP prover with RAMS can attain orders of magnitude speedup over the naive one without RAMS, and can achieve good performance for non-Horn problems on a non-shared memory multiprocessor, Multi-PSI.
Beschreibung:19 S.