Title :
An Upper Bound on Checking Test Complexity for Almost All Cographs
Author :
Zubkov, Oleg V. ; Chistikov, Dmitry V. ; Voronenko, Andrey A.
Author_Institution :
Dept. of Comput. Sci., East-Siberian State Acad. of Educ., Irkutsk, Russia
Abstract :
The concept of a checking test is of prime interest to the study of a variant of exact identification problem, in which the learner is given a hint about the unknown object. A graph F is said to be a checking test for a co graph G iff for any other co graph H there exists an edge in F distinguishing G and H, that is, contained in exactly one of the graphs G and H. It is known that for any co graph G there exists a unique irredundant checking test, the number of edges in which is called the checking test complexity of G. We show that almost all co graphs on n vertices have relatively small checking test complexity of O(n log n). Using this result, we obtain an upper bound on the checking test complexity of almost all read-once Boolean functions over the basis of disjunction and parity functions.
Keywords :
Boolean functions; computational complexity; graph theory; cographs; identification problem; read-once Boolean functions; test complexity checking; Binary trees; Boolean functions; Complexity theory; Education; Polynomials; Upper bound; Vectors; checking test; cograph; complexity; equivalence query; read-once Boolean function; teaching;
Conference_Titel :
Symbolic and Numeric Algorithms for Scientific Computing (SYNASC), 2011 13th International Symposium on
Conference_Location :
Timisoara
Print_ISBN :
978-1-4673-0207-4
DOI :
10.1109/SYNASC.2011.44