Impact of Delays on Constrained Online Convex Optimization

Xuanyu Cao, Junshan Zhang, H. Vincent Poor

Research output: Chapter in Book/Report/Conference proceedingConference contribution

4 Scopus citations

Abstract

In this paper, we study constrained online convex optimization (OCO) in the presence of feedback delays. The loss/constraint functions vary with time and their feedback information is revealed to the decision maker with delays, which arise naturally in many applications due to the latency associated with computation and communication. The effects of delays are not captured by standard OCO, where feedback information is disclosed to the decision maker immediately after a decision is made. We develop a modified online saddle point algorithm for constrained OCO with feedback delays. Sublinear regret and sublinear constraint violation bounds are established for the proposed algorithm and the impact of delays on the performance of the algorithm is highlighted.

Original languageEnglish (US)
Title of host publicationConference Record - 53rd Asilomar Conference on Circuits, Systems and Computers, ACSSC 2019
EditorsMichael B. Matthews
PublisherIEEE Computer Society
Pages1578-1581
Number of pages4
ISBN (Electronic)9781728143002
DOIs
StatePublished - Nov 2019
Externally publishedYes
Event53rd Asilomar Conference on Circuits, Systems and Computers, ACSSC 2019 - Pacific Grove, United States
Duration: Nov 3 2019Nov 6 2019

Publication series

NameConference Record - Asilomar Conference on Signals, Systems and Computers
Volume2019-November
ISSN (Print)1058-6393

Conference

Conference53rd Asilomar Conference on Circuits, Systems and Computers, ACSSC 2019
Country/TerritoryUnited States
CityPacific Grove
Period11/3/1911/6/19

All Science Journal Classification (ASJC) codes

  • Signal Processing
  • Computer Networks and Communications

Keywords

  • Online convex optimization
  • constrained optimization
  • feedback delay

Fingerprint

Dive into the research topics of 'Impact of Delays on Constrained Online Convex Optimization'. Together they form a unique fingerprint.

Cite this