Can Social Learning Increase Learning Speed, Performance or Both?

J.V. Heinerman, J. Stork, M.A. Rebolledo Coy, J.G. Hubert, A.E. Eiben, Thomas Bartz-Beielstein, Evert Haasdijk

Research output: Contribution to ConferencePaperOther research output

Abstract

Social learning enables multiple robots to share learned experiences while completing a task. The literature offers contradicting examples of its benefits; robots trained with social learning reach a higher performance, an increased learning speed, or both, compared to their individual learning counterparts. No general explanation has been advanced for the difference in observations, which make the results highly dependent on the particular system and parameter setting. In this research, we show that even within one system, the observed advantages of social learning can vary between parameter settings. Using Evolutionary Robotics, we train robots individually in a foraging task. We compare the performance of 50 parameter instances of the evolutionary algorithm obtained by a definitive screening design. We apply social learning in groups of two and four robots to the parameter settings that lead to the best and median performance. Our results show that the observed advantages of social learning differ highly between parameter settings but in general, median quality parameter settings experience more benefit from social learning. These results serve as a reminder that tuning of the parameters should not be left as an afterthought because they can drastically impact the conclusions on the advantages of social learning. Additionally, these results suggest that social learning reduces the sensitivity of the learning process to the choice of parameters.
LanguageEnglish
Pages200-207
Number of pages8
Publication statusPublished - 2017

Fingerprint

Robots
Evolutionary algorithms
Screening
Robotics
Tuning

Cite this

@conference{e8faa11b58dc48a89e7852449c3eb07d,
title = "Can Social Learning Increase Learning Speed, Performance or Both?",
abstract = "Social learning enables multiple robots to share learned experiences while completing a task. The literature offers contradicting examples of its benefits; robots trained with social learning reach a higher performance, an increased learning speed, or both, compared to their individual learning counterparts. No general explanation has been advanced for the difference in observations, which make the results highly dependent on the particular system and parameter setting. In this research, we show that even within one system, the observed advantages of social learning can vary between parameter settings. Using Evolutionary Robotics, we train robots individually in a foraging task. We compare the performance of 50 parameter instances of the evolutionary algorithm obtained by a definitive screening design. We apply social learning in groups of two and four robots to the parameter settings that lead to the best and median performance. Our results show that the observed advantages of social learning differ highly between parameter settings but in general, median quality parameter settings experience more benefit from social learning. These results serve as a reminder that tuning of the parameters should not be left as an afterthought because they can drastically impact the conclusions on the advantages of social learning. Additionally, these results suggest that social learning reduces the sensitivity of the learning process to the choice of parameters.",
author = "J.V. Heinerman and J. Stork and {Rebolledo Coy}, M.A. and J.G. Hubert and A.E. Eiben and Thomas Bartz-Beielstein and Evert Haasdijk",
year = "2017",
language = "English",
pages = "200--207",

}

Can Social Learning Increase Learning Speed, Performance or Both? / Heinerman, J.V.; Stork, J.; Rebolledo Coy, M.A.; Hubert, J.G.; Eiben, A.E.; Bartz-Beielstein, Thomas; Haasdijk, Evert.

2017. 200-207.

Research output: Contribution to ConferencePaperOther research output

TY - CONF

T1 - Can Social Learning Increase Learning Speed, Performance or Both?

AU - Heinerman, J.V.

AU - Stork, J.

AU - Rebolledo Coy, M.A.

AU - Hubert, J.G.

AU - Eiben, A.E.

AU - Bartz-Beielstein, Thomas

AU - Haasdijk, Evert

PY - 2017

Y1 - 2017

N2 - Social learning enables multiple robots to share learned experiences while completing a task. The literature offers contradicting examples of its benefits; robots trained with social learning reach a higher performance, an increased learning speed, or both, compared to their individual learning counterparts. No general explanation has been advanced for the difference in observations, which make the results highly dependent on the particular system and parameter setting. In this research, we show that even within one system, the observed advantages of social learning can vary between parameter settings. Using Evolutionary Robotics, we train robots individually in a foraging task. We compare the performance of 50 parameter instances of the evolutionary algorithm obtained by a definitive screening design. We apply social learning in groups of two and four robots to the parameter settings that lead to the best and median performance. Our results show that the observed advantages of social learning differ highly between parameter settings but in general, median quality parameter settings experience more benefit from social learning. These results serve as a reminder that tuning of the parameters should not be left as an afterthought because they can drastically impact the conclusions on the advantages of social learning. Additionally, these results suggest that social learning reduces the sensitivity of the learning process to the choice of parameters.

AB - Social learning enables multiple robots to share learned experiences while completing a task. The literature offers contradicting examples of its benefits; robots trained with social learning reach a higher performance, an increased learning speed, or both, compared to their individual learning counterparts. No general explanation has been advanced for the difference in observations, which make the results highly dependent on the particular system and parameter setting. In this research, we show that even within one system, the observed advantages of social learning can vary between parameter settings. Using Evolutionary Robotics, we train robots individually in a foraging task. We compare the performance of 50 parameter instances of the evolutionary algorithm obtained by a definitive screening design. We apply social learning in groups of two and four robots to the parameter settings that lead to the best and median performance. Our results show that the observed advantages of social learning differ highly between parameter settings but in general, median quality parameter settings experience more benefit from social learning. These results serve as a reminder that tuning of the parameters should not be left as an afterthought because they can drastically impact the conclusions on the advantages of social learning. Additionally, these results suggest that social learning reduces the sensitivity of the learning process to the choice of parameters.

M3 - Paper

SP - 200

EP - 207

ER -