Papers
Topics
Authors
Recent
Gemini 2.5 Flash
Gemini 2.5 Flash
134 tokens/sec
GPT-4o
10 tokens/sec
Gemini 2.5 Pro Pro
47 tokens/sec
o3 Pro
4 tokens/sec
GPT-4.1 Pro
38 tokens/sec
DeepSeek R1 via Azure Pro
28 tokens/sec
2000 character limit reached

Unbiased Estimating Equation on Inverse Divergence and Its Conditions (2404.16519v1)

Published 25 Apr 2024 in cs.IT, cs.LG, math.IT, math.ST, and stat.TH

Abstract: This paper focuses on the Bregman divergence defined by the reciprocal function, called the inverse divergence. For the loss function defined by the monotonically increasing function $f$ and inverse divergence, the conditions for the statistical model and function $f$ under which the estimating equation is unbiased are clarified. Specifically, we characterize two types of statistical models, an inverse Gaussian type and a mixture of generalized inverse Gaussian type distributions, to show that the conditions for the function $f$ are different for each model. We also define Bregman divergence as a linear sum over the dimensions of the inverse divergence and extend the results to the multi-dimensional case.

Definition Search Book Streamline Icon: https://streamlinehq.com
References (15)
  1. A. Basu, I. R. Harris, N. L. Hjort, and M. C. Jones, “Robust and efficient estimation by minimising a density power divergence,” Biometrika, vol. 85, no. 3, pp. 549–559, 1998.
  2. T. Mukherjee, A. Mandal, and A. Basu, “The B-exponential divergence and its generalizations with applications to parametric estimation,” Statistical Methods & Applications, vol. 28, no. 2, pp. 241–257, 2019.
  3. S. Roy, K. Chakraborty, S. Bhadra, and A. Basu, “Density power downweighting and robust inference: Some new strategies,” Journal of Mathematics and Statistics, vol. 15, pp. 333–353, 2019.
  4. P. Singh, A. Mandal, and A. Basu, “Robust inference using the exponential-polynomial divergence,” Journal of Statistical Theory and Practice, vol. 15, no. 2, pp. 1–22, 2021.
  5. N. Murata, T. Takenouchi, T. Kanamori, and S. Eguchi, “Information geometry of U𝑈{U}italic_U-boost and Bregman divergence,” Neural Computation, vol. 16, no. 7, pp. 1437–1481, 2004.
  6. H. Fujisawa and S. Eguchi, “Robust parameter estimation with a small bias against heavy contamination,” Journal of Multivariate Analysis, vol. 99, no. 9, pp. 2053–2081, 2008.
  7. H. Fujisawa, “Normalized estimating equation for robust parameter estimation,” Electronic Journal of Statistics, vol. 7, pp. 1587–1606, 2013.
  8. A. Okuno, “Minimizing robust density power-based divergences for general parametric density models,” Annals of the Institute of Statistical Mathematics, 2024, to be published.
  9. L. Yu, J. Song, Y. Song, and S. Ermon, “Pseudo-spherical contrastive divergence,” in Proc. Advances in Neural Information Processing Systems (NeurIPS), vol. 34, 2021, pp. 22 348–22 362.
  10. Y. Shkel and S. Verdú, “A coding theorem for f-separable distortion measures,” Entropy, vol. 20, no. 2, pp. 1–16, 2018.
  11. M. Kobayashi and K. Watanabe, “Generalized Dirichlet-process-means for f𝑓fitalic_f-separable distortion measures,” Neurocomputing, vol. 458, pp. 667–689, 2021.
  12. ——, “Unbiased estimating equation and latent bias under f𝑓fitalic_f-separable Bregman distortion measures,” IEEE Transactions on Information Theory, 2024, to be published.
  13. A. Sanhueza, V. Leiva, and N. Balakrishnan, “A new class of inverse Gaussian type distributions,” Metrika, vol. 68, no. 1, pp. 31–49, 2008.
  14. A. Banerjee, S. Merugu, I. S. Dhillon, and J. Ghosh, “Clustering with Bregman divergences,” Journal of Machine Learning Research, vol. 6, pp. 1705–1749, 2005.
  15. A. Cichocki and S. Amari, “Families of alpha- beta- and gamma- divergences: Flexible and robust measures of similarities,” Entropy, vol. 12, no. 6, pp. 1532–1568, 2010.

Summary

We haven't generated a summary for this paper yet.