Abstract
A majority of studies on inductive inference of formal languages and models of logic programming have mainly used Gold's identification in the limit as a correct inference criterion. In this criterion, we can not decide in general whether the inference terminates or not, and the results of the inference necessarily involve some risks. In this paper, we deal with finite identification for a class of recursive languages. The inference machine produces a unique guess just once when it is convinced the termination of the inference, and the results do not. involve any risks at all. We present necessary and sufficient conditions for a class of recursive languages to be finitely identifiable from positive or complete data. We also present some classes of recursive languages that are finitely identifiable from positive or complete data.
Preview
Unable to display preview. Download preview PDF.
References
Angluin, D.: Inductive inference of formal languages from positive data, Information and Control 45 (1980), 117–135
Angluin, D.: Finding patterns common to a set of strings, Proc. 11th Annual Symposium on Theory of Computing (1979), 130–141
Angluin, D., Smith, C.H.: Inductive inference: theory and methods, ACM Computing Surveys 15 No. 3 (1983), 237–269
Freivald R.V., Wiehagen, R.: Inductive inference with additional information, Elektron. Informationsverarb. Kybern. (EIK) 15 (1979), 179–185
Gold, E.M.: Language identification in the limit, Information and Control, 10 (1967), 447–474
Jantke, K.P., Beick, H.-R.: Combining postulates of naturalness in inductive inference, Elektron. Informationsverarb. Kybern. (EIK) 17 (1981), 465–484
Klette, R., Wiehagen, R.: Research in the theory of inductive inference by GDR mathematicians — a survey, Information Sciences 22 (1980), 149–169
Lange, S., Zeugmann, T.: On the power of monotonic language learning, GOSLER-Report 05/92, Fachbereich Mathematik und Informatik, TH Leipzig (1992)
Lange, S., Zeuginann, T.: Types of monotonic language learning and their characterization, to appear in Proc. 5th Workshop on Comput. Learning Theory (1992)
Muggleton, S., Buntine, W.: Machine invention of first-order predicates by inverting resolution, Proc. 5th International Conference on Machine Learning (1988), 339–352
Mukouchi, Y.: Characterization of pattern languages, Proc. 2nd Workshop on Algorithmic Learning Theory (1991), 93–104
Mukouchi, Y.: Definite inductive inference as a successful identification criterion, RIFIS-TR-CS-52, Research Institute of Fundamental Information Science, Kyushu University, (1991)
Sato, M., Umayahara, K.: Inductive inferability for formal languages from positive data, Proc. 2nd Workshop on Algorithmic Learning Theory (1991), 84–92
Shapiro, E.Y.: Inductive inference of theories from facts, Technical Report 192, Department of Computer Science, Yale University, (1981)
Shinohara, T.: Inductive inference from positive data is powerful, Proc. 3rd Workshop on Comput. Learning Theory (1990), 97–110
Wright, K.: Identification of unions of languages drawn from an identifiable class, Proc. 2nd Workshop on Comput. Learning Theory (1989), 328–333
Author information
Authors and Affiliations
Editor information
Rights and permissions
Copyright information
© 1992 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Mukouchi, Y. (1992). Characterization of finite identification. In: Jantke, K.P. (eds) Analogical and Inductive Inference. AII 1992. Lecture Notes in Computer Science, vol 642. Springer, Berlin, Heidelberg. https://doi.org/10.1007/3-540-56004-1_18
Download citation
DOI: https://doi.org/10.1007/3-540-56004-1_18
Published:
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-56004-3
Online ISBN: 978-3-540-47339-8
eBook Packages: Springer Book Archive