Abstract
Recently, the widely applicable information criterion (WAIC) model selection method has been considered for reproducing and estimating a probability function from data in a learning system. The learning coefficient in Bayesian estimation serves to measure the learning efficiency in singular learning models, and has an important role in the WAIC method. Mathematically, the learning coefficient is the log canonical threshold of the relative entropy. In this paper, we consider the Vandermonde matrix-type singularity learning coefficients in statistical learning theory.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
H. Akaike, A new look at the statistical model identification. IEEE Trans. Autom. Control 9, 716–723 (1974)
M. Aoyagi, Log canonical threshold of Vandermonde matrix type singularities and generalization error of a three layered neural network. Int. J. Pure Appl. Math. 52, 177–204 (2009)
M. Aoyagi, Learning coefficient in Bayesian estimation of restricted Boltzmann machine. J. Algebraic Stat. 4 (1), 30–57 (2013)
M. Aoyagi, Consideration on singularities in learning theory and the learning coefficient. Entropy 15 (9), 3714–3733 (2013). doi:10.3390/e15093714
M. Aoyagi, K. Nagata, Learning coefficient of generalization error in Bayesian estimation and Vandermonde matrix type singularity. Neural Comput. 24,1569–1610 (2012)
I.N. Bernstein, The analytic continuation of generalized functions with respect to a parameter. Funct. Anal. Appl. 6 (6), 26–40 (1972)
J.E. Bjőrk, Rings of Differential Operators (North-Holland, Amsterdam, 1979)
W. Fulton, Introduction to Toric Varieties. Annals of Mathematics Studies (Princeton University Press, Princeton, NJ, 1993)
H. Hironaka, Resolution of singularities of an algebraic variety over a field of characteristic zero. Ann. Math 79, 109–326 (1964)
M. Kashiwara, B-functions and holonomic systems. Invent. Math. 38, 33–53 (1976)
J. Kollár, Singularities of Pairs, in Algebraic Geometry-Santa Cruz 1995, Series Proceedings of Symposia in Pure Mathematics, 9–29 July 1995, vol. 62 (American Mathematical Society, Providence, RI, 1997), pp. 221–287
S. Lin, Asymptotic approximation of marginal likelihood integrals (2010). arXiv:1003.5338v2
M. Mustata, Singularities of pairs via jet schemes. J. Am. Math. Soc. 15, 599–615 (2002)
T. Ohsawa, K. Takegoshi, On the extension of L 2 holomorphic functions. Math. Z. 195, 197–204 (1987)
D. Rusakov, D. Geiger, Asymptotic model selection for naive Bayesian networks. J. Mach. Learn. Res. 6, 1–35 (2005)
S. Watanabe, Algebraic Geometry and Statistical Learning Theory, vol. 25 (Cambridge University Press, New York, 2009)
P. Zwiernik, An asymptotic behavior of the marginal likelihood for general Markov models. J. Mach. Learn. Res. 12, 3283–3310 (2011)
Acknowledgements
This research was supported by the Ministry of Education, Culture, Sports, Science and Technology in Japan, Grant-in-Aid for Scientific Research 22540224.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2017 Springer International Publishing AG
About this paper
Cite this paper
Aoyagi, M. (2017). Learning Coefficients and Reproducing True Probability Functions in Learning Systems. In: Dang, P., Ku, M., Qian, T., Rodino, L. (eds) New Trends in Analysis and Interdisciplinary Applications. Trends in Mathematics(). Birkhäuser, Cham. https://doi.org/10.1007/978-3-319-48812-7_44
Download citation
DOI: https://doi.org/10.1007/978-3-319-48812-7_44
Published:
Publisher Name: Birkhäuser, Cham
Print ISBN: 978-3-319-48810-3
Online ISBN: 978-3-319-48812-7
eBook Packages: Mathematics and StatisticsMathematics and Statistics (R0)