Abstract
The induction of a minimal nondeterministic finite automaton (NFA) consistent with a given set of examples and counterexamples, which is known to be computationally hard, is discussed. The paper is an extension to the novel approach of transforming the problem of NFA induction into the integer nonlinear programming (INLP) problem. An improved formulation of the problem is proposed along with the two parallel algorithms to solve it. The methods for the distribution of tasks among processors along with distributed termination detection are presented. The experimental results for selected benchmarks are also reported.
Author supplied keywords
Cite
CITATION STYLE
Jastrzab, T., Czech, Z. J., & Wieczorek, W. (2016). Parallel induction of nondeterministic finite automata. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 9573, pp. 248–257). Springer Verlag. https://doi.org/10.1007/978-3-319-32149-3_24
Register to see more suggestions
Mendeley helps you to discover research relevant for your work.
