ITU

Evolvability Search: Directly Selecting for Evolvability in order to Study and Produce It

Research output: Conference Article in Proceeding or Book/Report chapterArticle in proceedingsResearchpeer-review

Standard

Evolvability Search: Directly Selecting for Evolvability in order to Study and Produce It. / Mengistu, Henok; Lehman, Joel Anthony; Clune, Jeff.

Proceedings of the Genetic and Evolutionary Computation Conference 2016: GECCO '16. Association for Computing Machinery, 2016. p. 141-148.

Research output: Conference Article in Proceeding or Book/Report chapterArticle in proceedingsResearchpeer-review

Harvard

Mengistu, H, Lehman, JA & Clune, J 2016, Evolvability Search: Directly Selecting for Evolvability in order to Study and Produce It. in Proceedings of the Genetic and Evolutionary Computation Conference 2016: GECCO '16. Association for Computing Machinery, pp. 141-148. https://doi.org/10.1145/2908812.2908838

APA

Mengistu, H., Lehman, J. A., & Clune, J. (2016). Evolvability Search: Directly Selecting for Evolvability in order to Study and Produce It. In Proceedings of the Genetic and Evolutionary Computation Conference 2016: GECCO '16 (pp. 141-148). Association for Computing Machinery. https://doi.org/10.1145/2908812.2908838

Vancouver

Mengistu H, Lehman JA, Clune J. Evolvability Search: Directly Selecting for Evolvability in order to Study and Produce It. In Proceedings of the Genetic and Evolutionary Computation Conference 2016: GECCO '16. Association for Computing Machinery. 2016. p. 141-148 https://doi.org/10.1145/2908812.2908838

Author

Mengistu, Henok ; Lehman, Joel Anthony ; Clune, Jeff. / Evolvability Search: Directly Selecting for Evolvability in order to Study and Produce It. Proceedings of the Genetic and Evolutionary Computation Conference 2016: GECCO '16. Association for Computing Machinery, 2016. pp. 141-148

Bibtex

@inproceedings{00b2a215995046b79eabef66c5efbe39,
title = "Evolvability Search: Directly Selecting for Evolvability in order to Study and Produce It",
abstract = "One hallmark of natural organisms is their significant evolvability, i.e.,their increased potential for further evolution. However, reproducing such evolvability in artificial evolution remains a challenge, which both reduces the performance of evolutionary algorithms and inhibits the study of evolvable digital phenotypes. Although some types of selection in evolutionary computation indirectly encourage evolvability, one unexplored possibility is to directly select for evolvability. To do so, we estimate an individual's future potential for diversity by calculating the behavioral diversity of its immediate offspring, and select organisms with increased offspring variation. While the technique is computationally expensive, we hypothesized that direct selection would better encourage evolvability than indirect methods. Experiments in two evolutionary robotics domains confirm this hypothesis: in both domains, such Evolvability Search produces solutions with higher evolvability than those produced with Novelty Search or traditional objective-based search algorithms. Further experiments demonstrate that the higher evolvability produced by Evolvability Search in a training environment also generalizes, producing higher evolvability in a new test environment without further selection. Overall, Evolvability Search enables generating evolvability more easily and directly, facilitating its study and understanding, and may inspire future practical algorithms that increase evolvability without significant computational overhead.",
author = "Henok Mengistu and Lehman, {Joel Anthony} and Jeff Clune",
year = "2016",
doi = "10.1145/2908812.2908838",
language = "English",
isbn = "978-1-4503-4206-3",
pages = "141--148",
booktitle = "Proceedings of the Genetic and Evolutionary Computation Conference 2016",
publisher = "Association for Computing Machinery",
address = "United States",

}

RIS

TY - GEN

T1 - Evolvability Search: Directly Selecting for Evolvability in order to Study and Produce It

AU - Mengistu, Henok

AU - Lehman, Joel Anthony

AU - Clune, Jeff

PY - 2016

Y1 - 2016

N2 - One hallmark of natural organisms is their significant evolvability, i.e.,their increased potential for further evolution. However, reproducing such evolvability in artificial evolution remains a challenge, which both reduces the performance of evolutionary algorithms and inhibits the study of evolvable digital phenotypes. Although some types of selection in evolutionary computation indirectly encourage evolvability, one unexplored possibility is to directly select for evolvability. To do so, we estimate an individual's future potential for diversity by calculating the behavioral diversity of its immediate offspring, and select organisms with increased offspring variation. While the technique is computationally expensive, we hypothesized that direct selection would better encourage evolvability than indirect methods. Experiments in two evolutionary robotics domains confirm this hypothesis: in both domains, such Evolvability Search produces solutions with higher evolvability than those produced with Novelty Search or traditional objective-based search algorithms. Further experiments demonstrate that the higher evolvability produced by Evolvability Search in a training environment also generalizes, producing higher evolvability in a new test environment without further selection. Overall, Evolvability Search enables generating evolvability more easily and directly, facilitating its study and understanding, and may inspire future practical algorithms that increase evolvability without significant computational overhead.

AB - One hallmark of natural organisms is their significant evolvability, i.e.,their increased potential for further evolution. However, reproducing such evolvability in artificial evolution remains a challenge, which both reduces the performance of evolutionary algorithms and inhibits the study of evolvable digital phenotypes. Although some types of selection in evolutionary computation indirectly encourage evolvability, one unexplored possibility is to directly select for evolvability. To do so, we estimate an individual's future potential for diversity by calculating the behavioral diversity of its immediate offspring, and select organisms with increased offspring variation. While the technique is computationally expensive, we hypothesized that direct selection would better encourage evolvability than indirect methods. Experiments in two evolutionary robotics domains confirm this hypothesis: in both domains, such Evolvability Search produces solutions with higher evolvability than those produced with Novelty Search or traditional objective-based search algorithms. Further experiments demonstrate that the higher evolvability produced by Evolvability Search in a training environment also generalizes, producing higher evolvability in a new test environment without further selection. Overall, Evolvability Search enables generating evolvability more easily and directly, facilitating its study and understanding, and may inspire future practical algorithms that increase evolvability without significant computational overhead.

U2 - 10.1145/2908812.2908838

DO - 10.1145/2908812.2908838

M3 - Article in proceedings

SN - 978-1-4503-4206-3

SP - 141

EP - 148

BT - Proceedings of the Genetic and Evolutionary Computation Conference 2016

PB - Association for Computing Machinery

ER -

ID: 81058585