Abstract
This paper proposes an efficient numerical integration formula to compute the normalizing constant of Fisher–Bingham distributions. This formula uses a numerical integration formula with the continuous Euler transform to a Fourier-type integral representation of the normalizing constant. As this method is fast and accurate, it can be applied to the calculation of the normalizing constant of high-dimensional Fisher–Bingham distributions. More precisely, the error decays exponentially with an increase in the integration points, and the computation cost increases linearly with the dimensions. In addition, this formula is useful for calculating the gradient and Hessian matrix of the normalizing constant. Therefore, we apply this formula to efficiently calculate the maximum likelihood estimation (MLE) of high-dimensional data. Finally, we apply the MLE to the hyperspherical variational auto-encoder (S-VAE), a deep-learning-based generative model that restricts the latent space to a unit hypersphere. We use the S-VAE trained with images of handwritten numbers to estimate the distributions of each label. This application is useful for adding new labels to the models.
Similar content being viewed by others
References
Davidson, T.R., Falorsi, L., De Cao, N., Kipf, T., Tomczak, J.M.: Hyperspherical variational auto-encoders. arXiv preprint arXiv:1804.00891 (2018)
Doersch, C.: Tutorial on variational autoencoders. arXiv preprint arXiv:1606.05908 (2016)
Kingma, D.P., Welling, M.: Auto-encoding variational bayes. arXiv preprint arXiv:1312.6114 (2013)
Kume, A., Sei, T.: On the exact maximum likelihood inference of Fisher-Bingham distributions using an adjusted holonomic gradient method. Stat. Comput. 28(4), 835–847 (2018)
Kume, A., Wood, A.T.: Saddlepoint approximations for the Bingham and Fisher-Bingham normalising constants. Biometrika 92(2), 465–476 (2005)
Ooura, T.: A continuous Euler transformation and its application to the Fourier transform of a slowly decaying function. J. Comput. Appl. Math. 130(1–2), 259–270 (2001)
Sei, T., Kume, A.: Calculating the normalising constant of the Bingham distribution on the sphere using the holonomic gradient method. Stat. Comput. 25(2), 321–332 (2015)
Tanaka, K.: Error control of a numerical formula for the Fourier transform by Ooura’s continuous Euler transform and fractional FFT. J. Comput. Appl. Math. 266, 73–86 (2014)
Acknowledgements
We are grateful to Taichi Kiwaki for providing access to a GPU and giving advice about VAE. We thank Kazuki Matoya for general discussion about the application of MLE to VAE. We would like to thank Shun Sato for providing some advice about numerical computation. This work was supported by all members in the mathematical informatics 3rd laboratory of the University of Tokyo. Finally, we appreciate Tomonari Sei, who joined our discussion and gave a lot of advice. Ken’ichiro Tanaka is supported by the grant-in-aid of Japan Society of the Promotion of Science with KAKENHI Grant Number 17K14241.
Author information
Authors and Affiliations
Corresponding author
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Supplementary Information
Below is the link to the electronic supplementary material.
Rights and permissions
About this article
Cite this article
Chen, Y., Tanaka, K. Maximum likelihood estimation of the Fisher–Bingham distribution via efficient calculation of its normalizing constant. Stat Comput 31, 40 (2021). https://doi.org/10.1007/s11222-021-10015-9
Received:
Accepted:
Published:
DOI: https://doi.org/10.1007/s11222-021-10015-9