Abstract
For all complexity measures in Kolmogorov complexity the effect discovered by P. Martin-Löf holds. For every infinite binary sequence there is a wide gap between the supremum and the infimum of the complexity of initial fragments of the sequence. It is assumed that that this inevitable gap is characteristic of Kolmogorov complexity, and it is caused by the highly abstract nature of the unrestricted Kolmogorov complexity.
We consider the complexity of inductive inference for recursively enumerable classes of total recursive functions. This object is considered as a rather simple object where no effects from highly abstract theories are likely to be met. Here, similar gaps were discovered. Moreover, the existence of these gaps is proved by an explicit use of the theorem by P. Martin-Löf.
In our paper, we study a modification of inductive inference complexity. The complexity is usually understood as the maximum of mindchanges over the functions defined by the first n indices of the numbering. Instead we consider the mindchange complexity as the maximum over the first n functions in the numbering (disregarding the repeated functions. Linear upper and lower bounds for the mindchange complexity are proved. However, the gap between bounds for all n and bounds for infinitely many n, remains.
Supported by Latvia Science Council Grant 96.0282
Supported by NSF Grant 9301339.
Supported by Auckland University Research Grants A18/XXXXX/62090/3414012, A18/XXXXX/62090/F3414022
Supported by the Internatonal Computer Science Institute, Berkeley, California, by the DFG grant KA 673/4-1, and by the ESPRIT BR Grants 7079 and ECUS030
Preview
Unable to display preview. Download preview PDF.
References
A. Ambainis and J.Smotrovs, Enumerable classes of total recursive functions: complexity of inductive inference. Proceedings of the AII'94, Lecture Notes in Computer Science, 872 (1994), pp. 10–25
J. Bārzdiņš. On the frequency solution of the membership problem for recursively enumerable sets. Trudi of V.A.Steklov Mathematics Institute, Moscow, 1973, pp. 133–160
L. Blum and M.Blum, Towards a mathematical theory of inductive inference, Information and Control, 28 (1975), pp.122–155
J. Bārziņš and R. Freivalds. On the prediction of general recursive functions. Soviet Math. Doklady, 13 (1972), pp.1224–1228.
J. Case, C. Smith, Comparison of identification criteria for machine inductive inference. Theoretical Computer Science, 25 (1983), pp.193–220
R. Freivalds, J. Bārzdiņš, K. Podnieks. Inductive inference of recursive functions: complexity bounds. Baltic Computer Science. Lecture Notes in Computer Science, 502 (1991), pp.111–155.
R. Freivalds and A.C. Hoffman. An inductive inference approach to classification. Journal of Experimental and Theoretical Artificial Inteligence, 6 (1994), pp.63–72.
R. Freivalds, C. H. Smith, The role of procrastination in machine learning. Information and Computation, vol. 107 (1993), pp. 237–271
E.M.Gold, Limiting recursion. Journal of Symbolic Logic, 30 (1965), pp.28–48
E.M.Gold, Language identification in the limit. Information and Control, 10 (1967), pp.447–474
Efim Kinber and Frank Stephan. Language learning from texts: mind changes, limited memory and monotonicity. Proc. 8th Annu. Conf. on Comput. Learning Theory,ACM Press, New York, NY, pp. 182–189, 1995
A.N. Kolmogorov, Three approaches to the quantitative definition of information. Problems of Information Transmission, vol. 1, No. 1, 1965, pp.1–7.
S. Lange, T. Zeugmann, Trading monotonicity demands versus mind changes. Proceedings of EuroCOLT'95, Lecture Notes in Computer Science, vol. 904 (1995), pp. 125–139
M. Li, P. Vitanyi, Introduction to Kolmogorov Complexity and Its Applications. Springer-Verlag, 1993
P. Martin-Löf. On the definition of random sequence. Information and Control, 9 (1966), pp.602–619
H. Rogers, Jr. Gödel numberings of partial recursive functions. Journal of Symbolic Logic, 23:331–341, 1958
H. Rogers, Jr. Theory of Recursive Functions and Effective Computability. McGraw-Hill, 1967. Reprinted, MIT Press, 1987
R. Soare, Recursively Enumerable Sets and Degrees. Springer-Verlag, 1987 *** DIRECT SUPPORT *** A0008157 00005
Author information
Authors and Affiliations
Editor information
Rights and permissions
Copyright information
© 1997 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Ambainis, A. et al. (1997). Effects of Kolmogorov complexity present in inductive inference as well. In: Li, M., Maruoka, A. (eds) Algorithmic Learning Theory. ALT 1997. Lecture Notes in Computer Science, vol 1316. Springer, Berlin, Heidelberg. https://doi.org/10.1007/3-540-63577-7_47
Download citation
DOI: https://doi.org/10.1007/3-540-63577-7_47
Published:
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-63577-2
Online ISBN: 978-3-540-69602-5
eBook Packages: Springer Book Archive