Dynamic stochastic optimization

Craig Wilson, Venugopal Veeravalli, Angelia Nedich

Research output: Contribution to journalArticle

3 Scopus citations

Abstract

A framework for sequentially solving stochastic optimization problems with stochastic gradient descent is introduced. Two tracking criteria are considered, one based on being accurate with respect to the mean trajectory and the other based on being accurate in high probability (IHP). An off-line optimization problem is solved to find the constant step size and number of iterations to achieve the desired tracking accuracy. Simulations are used to confirm that this approach provides the desired tracking accuracy.

Original languageEnglish (US)
Article number7039377
Pages (from-to)173-178
Number of pages6
JournalUnknown Journal
Volume2015-February
Issue numberFebruary
DOIs
StatePublished - 2014
Externally publishedYes

Keywords

  • adaptive optimization
  • gradient methods
  • stochastic optimization

ASJC Scopus subject areas

  • Control and Systems Engineering
  • Modeling and Simulation
  • Control and Optimization

Fingerprint Dive into the research topics of 'Dynamic stochastic optimization'. Together they form a unique fingerprint.

  • Cite this

    Wilson, C., Veeravalli, V., & Nedich, A. (2014). Dynamic stochastic optimization. Unknown Journal, 2015-February(February), 173-178. [7039377]. https://doi.org/10.1109/CDC.2014.7039377