Learning Regular Omega Languages
We provide an algorithm for learning an unknown regular set of infinite words, using membership and equivalence queries. Three variations of the algorithm learn three different canonical representations of omega regular languages, using the notion of families of dfas. One is of size similar to L $, a dfa representation recently learned using L* . The second is based on the syntactic forc, introduced in . The third is introduced herein.We show that the second can be exponentially smaller than the first, and the third is at most as large as the first two, with up to a quadratic saving with respect to the second.
KeywordsRegular Language Canonical Representation Acceptance Condition Membership Query Equivalence Query
Unable to display preview. Download preview PDF.
- 1.Alur, R., Cerný, P., Madhusudan, P., Nam, W.: Synthesis of interface specifications for Java classes. In: POPL, pp. 98–109 (2005)Google Scholar
- 8.Jayasrirani, M., Humrosia Begam, M., Thomas, D.G., Emerald, J.D.: Learning of bi-languages from factors. Machine Learning Research Jour. 21, 139–144 (2012)Google Scholar
- 10.Klarlund, N.: A homomorphism concept for omega-regularity. In: 8th Inter. Conf. on Computer Science Logic (CSL), pp. 471–485 (1994)Google Scholar
- 16.Michel, M.: Complementation is much more difficult with automata on infinite words. In: Manuscript, CNET (1988)Google Scholar
- 19.Saoudi, A., Yokomori, T.: Learning local and recognizable omega-languages and monadic logic programs. In: EUROCOLT. LNCS, vol. 1121, pp. 50–59. Springer (1993)Google Scholar
- 21.Trakhtenbrot, B.: Finite automata and monadic second order logic. Siberian Math. J., 103–131 (1962)Google Scholar