arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1202.5665eess.SYcs.ITcs.SYmath.IT
收藏

基于q-高斯分布的平滑泛函随机优化算法

q-Gaussian based Smoothed Functional Algorithm for Stochastic Optimization

Indian Institute of Science, Bangalore(印度科学学院)

Debarghya Ghoshdastidar, Ambedkar Dukkipati, Shalabh Bhatnagar

AI总结:

本文提出一种使用q-高斯分布进行梯度估计的平滑泛函方法,并基于此设计优化算法,证明了收敛性,并在排队模型上验证了性能。

摘要:

q-高斯分布源于在某些约束下最大化香农熵的某种推广。q-高斯分布的重要性在于它们表现出幂律行为,并且也推广了高斯分布。在本文中,我们提出了一种使用q-高斯分布的平滑泛函(SF)梯度估计方案,并基于上述方案提出了一种优化算法。给出了算法的收敛性结果。通过排队模型上的仿真结果展示了所提算法的性能。

英文摘要:

The q-Gaussian distribution results from maximizing certain generalizations of Shannon entropy under some constraints. The importance of q-Gaussian distributions stems from the fact that they exhibit power-law behavior, and also generalize Gaussian distributions. In this paper, we propose a Smoothed Functional (SF) scheme for gradient estimation using q-Gaussian distribution, and also propose an algorithm for optimization based on the above scheme. Convergence results of the algorithm are presented. Performance of the proposed algorithm is shown by simulation results on a queuing model.

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。

↑