Multidimensional stochastic approximation: Adaptive algorithms and applications

Mark Broadie, Deniz M. Cicek, Assaf Zeevi

Research output: Contribution to journalArticlepeer-review

Abstract

We consider prototypical sequential stochastic optimization methods of Robbins-Monro (RM), Kiefer-Wolfowitz (KW), and Simultaneous Perturbations Stochastic Approximation (SPSA) varieties and propose adaptive modifications for multidimensional applications. These adaptive versions dynamically scale and shift the tuning sequences to better match the characteristics of the unknown underlying function, as well as the noise level. We test our algorithms on a variety of representative applications in inventory management, health care, revenue management, supply chain management, financial engineering, and queueing theory.

Original languageEnglish
Article number6
JournalACM Transactions on Modeling and Computer Simulation
Volume24
Issue number1
DOIs
StatePublished - Jan 2014
Externally publishedYes

Keywords

  • Adaptive algorithms
  • Algorithms
  • G.1.6. [optimization]: stochastic programming
  • G.4. [mathematical software]: algorithm design and analysis
  • I.6.3. [simulation and modelling]: applications
  • Kiefer-Wolfowitz
  • Numerical examples
  • Performance
  • Robbins-Monro
  • Simultaneous perturbations stochastic approximation
  • Stochastic approximations

Fingerprint

Dive into the research topics of 'Multidimensional stochastic approximation: Adaptive algorithms and applications'. Together they form a unique fingerprint.

Cite this