Momentum-Space Renormalization Group Transformation in Bayesian Image Modeling by Gaussian Graphical Model

03/20/2018 ∙ by Kazuyuki Tanaka, et al. ∙ Tohoku University 0

A new Bayesian modeling method is proposed by combining the maximization of the marginal likelihood with a momentum-space renormalization group transformation for Gaussian graphical models. Moreover, we present a scheme for computint the statistical averages of hyperparameters and mean square errors in our proposed method based on a momentumspace renormalization transformation.



There are no comments yet.


page 5

This week in AI

Get the week's most popular data science and artificial intelligence research sent straight to your inbox every Saturday.


This work was partly supported by the JST-CREST (No.JPMJCR1402) for Japan Science and Technology Agency and the JSPS KAKENHI Grant (No.25120009 and No.15K20870).


  • [1] D. Koller and N. Friedman: Probabilistic Graphical Models: Principles and Techniques (MIT Press, 2009).
  • [2] H. Nishimori: Statistical Physics of Spin Glass and Information Processing —Introduction— (Oxford University Press, Oxford, UK, 2001)
  • [3] M. Mézard and A. Montanari: Information, Physics and Computation (Oxford University Press, New York, USA, 2009)
  • [4] T. Kuwatani, K. Nagata, M. Okada, M. Toriumi: Phys. Rev. E, 90 (2014) 042137.
  • [5] T. Kuwatani, K. Nagata, M. Okada, M. Toriumi: Earth, Planets and Space, 66 (2014) 5.
  • [6] S. Kataoka, M. Yasuda, C. Furtlehner and K. Tanaka: Inverse Problems, 30 (2014) 025003.
  • [7] K. Tanaka, S. Kataoka, M. Yasuda and M. Ohzeki: J. Phys. Soc. Jpn 84 (2015) 045001.
  • [8] D. J. Amit and V. Martin-Mayor: Field Theory, the Renormalization Group, and Critical Phenomena: Graphs to Computers (World Scientific, 2005).