What are dynamic optimization problems?

Haobo Fu, Peter R. Lewis, Bernhard Sendhoff, Ke Tang, Xin Yao

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

Abstract

Dynamic Optimization Problems (DOPs) have been widely studied using Evolutionary Algorithms (EAs). Yet, a clear and rigorous definition of DOPs is lacking in the Evolutionary Dynamic Optimization (EDO) community. In this paper, we propose a unified definition of DOPs based on the idea of multiple-decision-making discussed in the Reinforcement Learning (RL) community. We draw a connection between EDO and RL by arguing that both of them are studying DOPs according to our definition of DOPs. We point out that existing EDO or RL research has been mainly focused on some types of DOPs. A conceptualized benchmark problem, which is aimed at the systematic study of various DOPs, is then developed. Some interesting experimental studies on the benchmark reveal that EDO and RL methods are specialized in certain types of DOPs and more importantly new algorithms for DOPs can be developed by combining the strength of both EDO and RL methods.

Original languageEnglish
Title of host publicationProceedings of the 2014 IEEE Congress on Evolutionary Computation, CEC 2014
PublisherIEEE
Pages1550-1557
Number of pages8
ISBN (Print)978-1-4799-6626-4
DOIs
Publication statusPublished - 2014
Event2014 IEEE Congress on Evolutionary Computation - Beijing, China
Duration: 6 Jul 201411 Jul 2014

Congress

Congress2014 IEEE Congress on Evolutionary Computation
Abbreviated titleCEC 2014
CountryChina
CityBeijing
Period6/07/1411/07/14

Fingerprint

Dynamic Optimization Problems
Evolutionary Dynamics
Dynamic Optimization
Evolutionary Optimization
Reinforcement Learning
Reinforcement learning
Benchmark
Evolutionary Algorithms
Experimental Study
Decision Making
Evolutionary algorithms

Bibliographical note

© 2014 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works.

Cite this

Fu, H., Lewis, P. R., Sendhoff, B., Tang, K., & Yao, X. (2014). What are dynamic optimization problems? In Proceedings of the 2014 IEEE Congress on Evolutionary Computation, CEC 2014 (pp. 1550-1557). IEEE. https://doi.org/10.1109/CEC.2014.6900316
Fu, Haobo ; Lewis, Peter R. ; Sendhoff, Bernhard ; Tang, Ke ; Yao, Xin. / What are dynamic optimization problems?. Proceedings of the 2014 IEEE Congress on Evolutionary Computation, CEC 2014. IEEE, 2014. pp. 1550-1557
@inproceedings{a06c3112d1bf4479b7d59f44b7398aff,
title = "What are dynamic optimization problems?",
abstract = "Dynamic Optimization Problems (DOPs) have been widely studied using Evolutionary Algorithms (EAs). Yet, a clear and rigorous definition of DOPs is lacking in the Evolutionary Dynamic Optimization (EDO) community. In this paper, we propose a unified definition of DOPs based on the idea of multiple-decision-making discussed in the Reinforcement Learning (RL) community. We draw a connection between EDO and RL by arguing that both of them are studying DOPs according to our definition of DOPs. We point out that existing EDO or RL research has been mainly focused on some types of DOPs. A conceptualized benchmark problem, which is aimed at the systematic study of various DOPs, is then developed. Some interesting experimental studies on the benchmark reveal that EDO and RL methods are specialized in certain types of DOPs and more importantly new algorithms for DOPs can be developed by combining the strength of both EDO and RL methods.",
author = "Haobo Fu and Lewis, {Peter R.} and Bernhard Sendhoff and Ke Tang and Xin Yao",
note = "{\circledC} 2014 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works.",
year = "2014",
doi = "10.1109/CEC.2014.6900316",
language = "English",
isbn = "978-1-4799-6626-4",
pages = "1550--1557",
booktitle = "Proceedings of the 2014 IEEE Congress on Evolutionary Computation, CEC 2014",
publisher = "IEEE",
address = "United States",

}

Fu, H, Lewis, PR, Sendhoff, B, Tang, K & Yao, X 2014, What are dynamic optimization problems? in Proceedings of the 2014 IEEE Congress on Evolutionary Computation, CEC 2014. IEEE, pp. 1550-1557, 2014 IEEE Congress on Evolutionary Computation, Beijing, China, 6/07/14. https://doi.org/10.1109/CEC.2014.6900316

What are dynamic optimization problems? / Fu, Haobo; Lewis, Peter R.; Sendhoff, Bernhard; Tang, Ke; Yao, Xin.

Proceedings of the 2014 IEEE Congress on Evolutionary Computation, CEC 2014. IEEE, 2014. p. 1550-1557.

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

TY - GEN

T1 - What are dynamic optimization problems?

AU - Fu, Haobo

AU - Lewis, Peter R.

AU - Sendhoff, Bernhard

AU - Tang, Ke

AU - Yao, Xin

N1 - © 2014 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works.

PY - 2014

Y1 - 2014

N2 - Dynamic Optimization Problems (DOPs) have been widely studied using Evolutionary Algorithms (EAs). Yet, a clear and rigorous definition of DOPs is lacking in the Evolutionary Dynamic Optimization (EDO) community. In this paper, we propose a unified definition of DOPs based on the idea of multiple-decision-making discussed in the Reinforcement Learning (RL) community. We draw a connection between EDO and RL by arguing that both of them are studying DOPs according to our definition of DOPs. We point out that existing EDO or RL research has been mainly focused on some types of DOPs. A conceptualized benchmark problem, which is aimed at the systematic study of various DOPs, is then developed. Some interesting experimental studies on the benchmark reveal that EDO and RL methods are specialized in certain types of DOPs and more importantly new algorithms for DOPs can be developed by combining the strength of both EDO and RL methods.

AB - Dynamic Optimization Problems (DOPs) have been widely studied using Evolutionary Algorithms (EAs). Yet, a clear and rigorous definition of DOPs is lacking in the Evolutionary Dynamic Optimization (EDO) community. In this paper, we propose a unified definition of DOPs based on the idea of multiple-decision-making discussed in the Reinforcement Learning (RL) community. We draw a connection between EDO and RL by arguing that both of them are studying DOPs according to our definition of DOPs. We point out that existing EDO or RL research has been mainly focused on some types of DOPs. A conceptualized benchmark problem, which is aimed at the systematic study of various DOPs, is then developed. Some interesting experimental studies on the benchmark reveal that EDO and RL methods are specialized in certain types of DOPs and more importantly new algorithms for DOPs can be developed by combining the strength of both EDO and RL methods.

UR - http://www.scopus.com/inward/record.url?scp=84908587586&partnerID=8YFLogxK

U2 - 10.1109/CEC.2014.6900316

DO - 10.1109/CEC.2014.6900316

M3 - Conference contribution

SN - 978-1-4799-6626-4

SP - 1550

EP - 1557

BT - Proceedings of the 2014 IEEE Congress on Evolutionary Computation, CEC 2014

PB - IEEE

ER -

Fu H, Lewis PR, Sendhoff B, Tang K, Yao X. What are dynamic optimization problems? In Proceedings of the 2014 IEEE Congress on Evolutionary Computation, CEC 2014. IEEE. 2014. p. 1550-1557 https://doi.org/10.1109/CEC.2014.6900316