Computational study on effectiveness of knowledge transfer in dynamic multi-objective optimisation

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

Standard

Computational study on effectiveness of knowledge transfer in dynamic multi-objective optimisation. / Ruan, Gan; Minku, Leandro; Menzel, Stefan; Sendhoff, Bernhard; Yao, Xin.

2020 IEEE Congress on Evolutionary Computation (CEC). IEEE Computer Society Press, 2020. p. 1-8.

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

Harvard

Ruan, G, Minku, L, Menzel, S, Sendhoff, B & Yao, X 2020, Computational study on effectiveness of knowledge transfer in dynamic multi-objective optimisation. in 2020 IEEE Congress on Evolutionary Computation (CEC). IEEE Computer Society Press, pp. 1-8, 2020 IEEE Congress on Evolutionary Computation (IEE CEC 2020), Glasgow, United Kingdom, 19/07/20. https://doi.org/10.1109/CEC48606.2020.9185907

APA

Ruan, G., Minku, L., Menzel, S., Sendhoff, B., & Yao, X. (2020). Computational study on effectiveness of knowledge transfer in dynamic multi-objective optimisation. In 2020 IEEE Congress on Evolutionary Computation (CEC) (pp. 1-8). IEEE Computer Society Press. https://doi.org/10.1109/CEC48606.2020.9185907

Vancouver

Ruan G, Minku L, Menzel S, Sendhoff B, Yao X. Computational study on effectiveness of knowledge transfer in dynamic multi-objective optimisation. In 2020 IEEE Congress on Evolutionary Computation (CEC). IEEE Computer Society Press. 2020. p. 1-8 https://doi.org/10.1109/CEC48606.2020.9185907

Author

Ruan, Gan ; Minku, Leandro ; Menzel, Stefan ; Sendhoff, Bernhard ; Yao, Xin. / Computational study on effectiveness of knowledge transfer in dynamic multi-objective optimisation. 2020 IEEE Congress on Evolutionary Computation (CEC). IEEE Computer Society Press, 2020. pp. 1-8

Bibtex

@inproceedings{e70a65a37f90492abf843e04148330f5,
title = "Computational study on effectiveness of knowledge transfer in dynamic multi-objective optimisation",
abstract = "Transfer learning has been used for solving multiple optimization and dynamic multi-objective optimization problems, since transfer learning is believed to be able to transfer useful information from one problem instance to help solving another related problem instance. This paper aims to study how effective transfer learning is in dynamic multi-objective optimization (DMO). Through computation time analysis of transfer learning, we show that the {\textquoteleft}inner{\textquoteright} optimization problem introduced by transfer learning is very time-consuming. In order to enhance the efficiency, two alternatives are computationally investigated on a number of dynamic bi- and tri-objective test problems. Experimental results have shown that the greatly enhanced efficiency does not result in much degeneration on the performance of transfer learning. Considering the high computational cost of transfer learning, it is likely that the original purpose of using transfer learning in DMO might be negated. In other words, the computation time saved in optimization is eaten up by computationally expensive transfer learning. As a result, there is less gain than expected in the overall computational efficiency. To verify this, experiments have been conducted, regarding using computational cost of transfer learning to optimize randomly generated solutions. The results have demonstrated that the convergence and diversity of final solutions generated from the random solutions are significantly better than those generated from transferred solutions under the same total computational budget.",
keywords = "Evolutionary Algorithms, Transfer Learning, Dynamic Multi-objective Optimization, Prediction-based Method",
author = "Gan Ruan and Leandro Minku and Stefan Menzel and Bernhard Sendhoff and Xin Yao",
year = "2020",
month = sep,
day = "3",
doi = "https://doi.org/10.1109/CEC48606.2020.9185907",
language = "English",
isbn = "978-1-7281-6930-9 (PoD)",
pages = "1--8",
booktitle = "2020 IEEE Congress on Evolutionary Computation (CEC)",
publisher = "IEEE Computer Society Press",
note = "2020 IEEE Congress on Evolutionary Computation (IEE CEC 2020) ; Conference date: 19-07-2020 Through 24-07-2020",

}

RIS

TY - GEN

T1 - Computational study on effectiveness of knowledge transfer in dynamic multi-objective optimisation

AU - Ruan, Gan

AU - Minku, Leandro

AU - Menzel, Stefan

AU - Sendhoff, Bernhard

AU - Yao, Xin

PY - 2020/9/3

Y1 - 2020/9/3

N2 - Transfer learning has been used for solving multiple optimization and dynamic multi-objective optimization problems, since transfer learning is believed to be able to transfer useful information from one problem instance to help solving another related problem instance. This paper aims to study how effective transfer learning is in dynamic multi-objective optimization (DMO). Through computation time analysis of transfer learning, we show that the ‘inner’ optimization problem introduced by transfer learning is very time-consuming. In order to enhance the efficiency, two alternatives are computationally investigated on a number of dynamic bi- and tri-objective test problems. Experimental results have shown that the greatly enhanced efficiency does not result in much degeneration on the performance of transfer learning. Considering the high computational cost of transfer learning, it is likely that the original purpose of using transfer learning in DMO might be negated. In other words, the computation time saved in optimization is eaten up by computationally expensive transfer learning. As a result, there is less gain than expected in the overall computational efficiency. To verify this, experiments have been conducted, regarding using computational cost of transfer learning to optimize randomly generated solutions. The results have demonstrated that the convergence and diversity of final solutions generated from the random solutions are significantly better than those generated from transferred solutions under the same total computational budget.

AB - Transfer learning has been used for solving multiple optimization and dynamic multi-objective optimization problems, since transfer learning is believed to be able to transfer useful information from one problem instance to help solving another related problem instance. This paper aims to study how effective transfer learning is in dynamic multi-objective optimization (DMO). Through computation time analysis of transfer learning, we show that the ‘inner’ optimization problem introduced by transfer learning is very time-consuming. In order to enhance the efficiency, two alternatives are computationally investigated on a number of dynamic bi- and tri-objective test problems. Experimental results have shown that the greatly enhanced efficiency does not result in much degeneration on the performance of transfer learning. Considering the high computational cost of transfer learning, it is likely that the original purpose of using transfer learning in DMO might be negated. In other words, the computation time saved in optimization is eaten up by computationally expensive transfer learning. As a result, there is less gain than expected in the overall computational efficiency. To verify this, experiments have been conducted, regarding using computational cost of transfer learning to optimize randomly generated solutions. The results have demonstrated that the convergence and diversity of final solutions generated from the random solutions are significantly better than those generated from transferred solutions under the same total computational budget.

KW - Evolutionary Algorithms

KW - Transfer Learning

KW - Dynamic Multi-objective Optimization

KW - Prediction-based Method

UR - https://wcci2020.org/

U2 - https://doi.org/10.1109/CEC48606.2020.9185907

DO - https://doi.org/10.1109/CEC48606.2020.9185907

M3 - Conference contribution

SN - 978-1-7281-6930-9 (PoD)

SP - 1

EP - 8

BT - 2020 IEEE Congress on Evolutionary Computation (CEC)

PB - IEEE Computer Society Press

T2 - 2020 IEEE Congress on Evolutionary Computation (IEE CEC 2020)

Y2 - 19 July 2020 through 24 July 2020

ER -