arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2407.06013cs.ITmath.IT
收藏

重访Arimoto-Blahut算法:基于近似的新分析

Revisit the Arimoto-Blahut algorithm: New Analysis with Approximation

Michail Fasoulakis, Konstantinos Varsos, Apostolos Traganitis

AI总结:

本文重新分析Arimoto-Blahut算法的收敛速率,证明对任意足够小的ε>0,收敛到ε-最优解的速率为逆指数O(log(m)/c^t),并据此推导出容量精确计算的新结果。

摘要:

根据Claude Shannon的开创性论文,离散无记忆信道容量的计算一直被认为是信息论中最重要和最基本的问题之一。近50年前,Arimoto和Blahut在各自的开创性论文中独立提出了相同的算法来解决这一问题。Arimoto-Blahut算法被证明在$t \to \infty$时收敛到信道容量,收敛速率的上界为$O\left(\log(m)/t\right)$,其中$m$是输入分布的大小。在唯一最优解位于输入概率单纯形内部的假设下,经过某个迭代$t^0$之后,收敛变为逆指数速率。最近,Nakagawa等人证明在某些特定情况下,收敛速率在最坏情况下为逆线性。在本文中,我们重新审视这一基础算法,重点分析其在容量近似方面的收敛速率。我们的主要结果表明,对于任意足够小的常数$\varepsilon > 0$,收敛到$\varepsilon$-最优解的速率为逆指数$O\left(\log(m)/c^t\right)$,其中$c > 1$为某个常数。基于这一结果,我们推导出关于容量计算的新颖且互补的结论,特别是在需要精确解的情况下。

英文摘要:

By the seminal paper of Claude Shannon \cite{Shannon48}, the computation of the capacity of a discrete memoryless channel has been considered as one of the most important and fundamental problems in Information Theory. Nearly 50 years ago, Arimoto and Blahut independently proposed identical algorithms to solve this problem in their seminal papers \cite{Arimoto1972AnAF, Blahut1972ComputationOC}. The Arimoto-Blahut algorithm was proven to converge to the capacity of the channel as $t \to \infty$, with a convergence rate upper bounded by $O\left(\log(m)/t\right)$, where $m$ is the size of the input distribution. Under the assumption that a unique optimal solution is in the interior of the input probability simplex, the convergence becomes inverse exponential after an iteration $t^0$ \cite{Arimoto1972AnAF}. More recently, it was demonstrated in \cite{Nakagawa2020AnalysisOT} that in certain specific cases, the convergence rate is at worst case inverse linear. In this paper, we revisit this fundamental algorithm analyzing its rate of convergence focusing on the approximation of the capacity. Our main result shows that the convergence rate to an $\varepsilon$-optimal solution, for any sufficiently small constant $\varepsilon > 0$, is inverse exponential $O\left(\log(m)/c^t\right)$, for some constant $c > 1$. Given this, we derive new and complementary results for the computation of capacity, particularly in cases where an exact solution is sought.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑