Abstract
Angluin (1980) showed that there is a consistent and conservative learner for the class of non-erasing pattern languages; however, most of these learners are NP-hard. In the current work, the complexity of consistent polynomial time learners for the class of non-erasing pattern languages is revisited, with the goal to close one gap left by Angluin, namely the question on what happens if the learner is not required to output each time a consistent pattern of maximum possible length. It is shown that consistent learners are non-uniformly W[1]-hard inside the fixed-parameter hierarchy of Downey and Fellows (1999), and that there is also a W[1]-complete such learner. Only when one requires that the learner is in addition both, conservative and class-preserving, then one can show that the learning task is NP-hard for certain alphabet-sizes.
Supported in part by NUS grant numbers R146-000-114-112 and R252-000-420-112.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Angluin, D.: Finding patterns common to a set of strings. J. of Comput. Syst. Sci. 21, 46–62 (1980)
Angluin, D.: Inductive inference of formal languages from positive data. Inform. Control 45, 117–135 (1980)
Case, J., Kötzing, T.: Dynamically delayed postdictive completeness and consistency in learning. In: Freund, Y., Györfi, L., Turán, G., Zeugmann, T. (eds) ALT 2008. LNAI, vol. 5254, pp. 389–403. Springer, Berlin (2008)
Case, J., Kötzing, T.: Difficulties in forcing fairness of polynomial time inductive inference. In: Gavaldà , R., Lugosi, G., Zeugmann, T., Zilles, S. (eds) ALT 2009. LNAI, vol. 5809, pp. 263–277. Springer, Berlin (2009)
Downey, R.G., Fellows, M.R.: Parameterized complexity. Monographs in computer science. Springer, Berlin (1999)
Erlebach, T., Rossmanith, P., Stadtherr, H., Steger, A., Zeugmann, T.: Learning one-variable pattern languages very efficiently on average, in parallel, and by asking queries. Theoret. Comput. Sci. 261, 119–156 (2001)
Gold, M.: Language identification in the limit. Inform. Control 10, 447–474 (1967)
Lange, S., Wiehagen, R.: Polynomial-time inference of arbitrary pattern languages. New Genera. Comput. 8, 361–370 (1991)
Lange, S., Zeugmann, T., Zilles, S.: Learning indexed families of recursive languages from positive data: a survey. Theoret. Comput. Sci. 397, 194–232 (2008)
Reidenbach, D.: A non-learnable class of E-pattern languages. Theoret. Comput. Sci. 350, 91–102 (2006)
Rossmanith, P., Zeugmann, T.: Stochastic finite learning of the pattern languages. Mach. Learn. 44, 67–91 (2001)
Shinohara, T.: Polynomial time inference of extended regular pattern languages. In: RIMS Symposium on Software Science and Engineering, Proc. LNCS, 147. pp. 115–127, Springer, Berlin (1983)
Shinohara, T., Arikawa, S.: Pattern inference. In: Algorithmic Learning for Knowledge-Based Systems, LNAI, 961. pp. 259–291, Springer, Berlin (1995)
Zeugmann, T.: Lange and Wiehagen’s pattern language learning algorithm: an average-case analysis with respect to its total learning time. Ann. Math. Artif. Intell. 23, 117–145 (1998)
Zeugmann, T., Lange, S.: A guided tour across the boundaries of learning recursive languages. In: Algorithmic Learning for Knowledge-Based Systems, LNAI, 961. pp. 190–258, Springer, Berlin (1995)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2011 Springer-Verlag London Limited
About this paper
Cite this paper
Stephan, F., Yoshinaka, R., Zeugmann, T. (2011). On the Parameterised Complexity of Learning Patterns. In: Gelenbe, E., Lent, R., Sakellari, G. (eds) Computer and Information Sciences II. Springer, London. https://doi.org/10.1007/978-1-4471-2155-8_35
Download citation
DOI: https://doi.org/10.1007/978-1-4471-2155-8_35
Published:
Publisher Name: Springer, London
Print ISBN: 978-1-4471-2154-1
Online ISBN: 978-1-4471-2155-8
eBook Packages: EngineeringEngineering (R0)