We gratefully acknowledge support from
the Simons Foundation and member institutions.
Full-text links:


Current browse context:


Change to browse by:

References & Citations

DBLP - CS Bibliography


(what is this?)
CiteULike logo BibSonomy logo Mendeley logo del.icio.us logo Digg logo Reddit logo ScienceWISE logo

Computer Science > Information Theory

Title: Analysis for the Slow Convergence in Arimoto Algorithm

Abstract: In this paper, we investigate the convergence speed of the Arimoto algorithm. By analyzing the Taylor expansion of the defining function of the Arimoto algorithm, we will clarify the conditions for the exponential or $1/N$ order convergence and calculate the convergence speed. We show that the convergence speed of the $1/N$ order is evaluated by the derivatives of the Kullback-Leibler divergence with respect to the input probabilities. The analysis for the convergence of the $1/N$ order is new in this paper. Based on the analysis, we will compare the convergence speed of the Arimoto algorithm with the theoretical values obtained in our theorems for several channel matrices.
Comments: 36 pages, 12 figures, submitted to IEEE Transactions on Information Theory
Subjects: Information Theory (cs.IT)
Cite as: arXiv:1809.00752 [cs.IT]
  (or arXiv:1809.00752v1 [cs.IT] for this version)

Submission history

From: Kenji Nakagawa [view email]
[v1] Tue, 4 Sep 2018 00:25:21 GMT (168kb)

Link back to: arXiv, form interface, contact.