English

A Probabilistic Sample Path Convergence Time Analysis of Drift-Plus-Penalty Algorithm for Stochastic Optimization

Optimization and Control 2016-12-20 v3

Abstract

This paper considers the problem of minimizing the time average of a controlled stochastic process subject to multiple time average constraints on other related processes. The probability distribution of the random events in the system is unknown to the controller. A typical application is time average power minimization subject to network throughput constraints for different users in a network with time varying channel conditions. We show that with probability at least 12δ1-2\delta, the classical drift-plus-penalty algorithm provides a sample path O(ε)\mathcal{O}(\varepsilon) approximation to optimality with a convergence time O(1ε2max{log21εlog2δ, log32δ})\mathcal{O}(\frac{1}{\varepsilon^2}\max\left\{\log^2\frac1\varepsilon\log\frac2\delta,~\log^3\frac2\delta\right\}), where ε>0\varepsilon>0 is a parameter related to the algorithm. When there is only one constraint, we further show that the convergence time can be improved to O(1ε2log21δ)\mathcal{O}\left(\frac{1}{\varepsilon^2}\log^2\frac1\delta\right).

Keywords

Cite

@article{arxiv.1510.02973,
  title  = {A Probabilistic Sample Path Convergence Time Analysis of Drift-Plus-Penalty Algorithm for Stochastic Optimization},
  author = {Xiaohan Wei and Hao Yu and Michael J. Neely},
  journal= {arXiv preprint arXiv:1510.02973},
  year   = {2016}
}

Comments

This is an updated version for IEEE Transactions on Automatic Control with changes highlighted

R2 v1 2026-06-22T11:17:23.967Z