Vipin Kumar, Dana S. Nau, Laveen N. Kanal

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


The authors present a general top-down procedure, GAO* , for searching acyclic AND/OR graphs with monotone cost functions. This procedure is a generalization of the AO* algorithm and can be viewed as a branch and bound procedure. Since monotone cost functions are very general, GAO* is applicable to a very large number of problems. For example, GAO* can be used to find the minimax value of a game tree. Furthermore, many existing game tree search procedures can be considered as special cases or slight variations of GAO* . The proof of correctness of GAO* is quite simple, which simplifies the correctness proof of AO* .

Original languageEnglish (US)
Title of host publicationProceedings - IEEE Computer Society's International Computer Software & Applications Conference
Number of pages5
ISBN (Print)0818606436
StatePublished - Dec 1 1985


Dive into the research topics of 'ALGORITHM.'. Together they form a unique fingerprint.

Cite this