Online Optimization Using Zeroth Order Oracles

Iman Shames; Daniel Selvaratnam; Jonathan H. Manton

doi:10.1109/LCSYS.2019.2921593

Online Optimization Using Zeroth Order Oracles

Iman Shames^*, Daniel Selvaratnam, Jonathan H. Manton

^*Corresponding author for this work

Research output: Contribution to journal › Article › peer-review

17 Citations (Scopus)

Abstract

This letter considers the iterative numerical optimization of time-varying cost functions where no gradient information is available at each iteration. In this case, the proposed algorithm estimates a directional derivative by finite differences. The main contributions are the derivation of error bounds for such algorithms and proposal of optimal algorithm parameter values, e.g., step-sizes, for strongly convex cost functions. The algorithm is applied to tackle a source localization problem using a sensing agent where the source actively evades the agent. Numerical examples are provided to illustrate the theoretical results.

Original language	English
Article number	8733040
Pages (from-to)	31-36
Number of pages	6
Journal	IEEE Control Systems Letters
Volume	4
Issue number	1
DOIs	https://doi.org/10.1109/LCSYS.2019.2921593
Publication status	Published - Jan 2020
Externally published	Yes

Access to Document

10.1109/LCSYS.2019.2921593

Cite this

@article{e84f01eb182f46209c78503b4d2e75d3,

title = "Online Optimization Using Zeroth Order Oracles",

abstract = "This letter considers the iterative numerical optimization of time-varying cost functions where no gradient information is available at each iteration. In this case, the proposed algorithm estimates a directional derivative by finite differences. The main contributions are the derivation of error bounds for such algorithms and proposal of optimal algorithm parameter values, e.g., step-sizes, for strongly convex cost functions. The algorithm is applied to tackle a source localization problem using a sensing agent where the source actively evades the agent. Numerical examples are provided to illustrate the theoretical results.",

keywords = "Optimization algorithms, autonomous systems, optimization",

author = "Iman Shames and Daniel Selvaratnam and Manton, {Jonathan H.}",

note = "Publisher Copyright: {\textcopyright} 2017 IEEE.",

year = "2020",

month = jan,

doi = "10.1109/LCSYS.2019.2921593",

language = "English",

volume = "4",

pages = "31--36",

journal = "IEEE Control Systems Letters",

issn = "2475-1456",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

number = "1",

}

TY - JOUR

T1 - Online Optimization Using Zeroth Order Oracles

AU - Shames, Iman

AU - Selvaratnam, Daniel

AU - Manton, Jonathan H.

PY - 2020/1

Y1 - 2020/1

N2 - This letter considers the iterative numerical optimization of time-varying cost functions where no gradient information is available at each iteration. In this case, the proposed algorithm estimates a directional derivative by finite differences. The main contributions are the derivation of error bounds for such algorithms and proposal of optimal algorithm parameter values, e.g., step-sizes, for strongly convex cost functions. The algorithm is applied to tackle a source localization problem using a sensing agent where the source actively evades the agent. Numerical examples are provided to illustrate the theoretical results.

AB - This letter considers the iterative numerical optimization of time-varying cost functions where no gradient information is available at each iteration. In this case, the proposed algorithm estimates a directional derivative by finite differences. The main contributions are the derivation of error bounds for such algorithms and proposal of optimal algorithm parameter values, e.g., step-sizes, for strongly convex cost functions. The algorithm is applied to tackle a source localization problem using a sensing agent where the source actively evades the agent. Numerical examples are provided to illustrate the theoretical results.

KW - Optimization algorithms

KW - autonomous systems

KW - optimization

UR - http://www.scopus.com/inward/record.url?scp=85067888745&partnerID=8YFLogxK

U2 - 10.1109/LCSYS.2019.2921593

DO - 10.1109/LCSYS.2019.2921593

M3 - Article

SN - 2475-1456

VL - 4

SP - 31

EP - 36

JO - IEEE Control Systems Letters

JF - IEEE Control Systems Letters

IS - 1

M1 - 8733040

ER -

Online Optimization Using Zeroth Order Oracles

Abstract

Access to Document

Other files and links

Fingerprint

Cite this