Interaction is More Beneficial in Complex Reinforcement Learning Problems than in Simple Ones

Research output: Contribution to book/anthology/report/proceedingArticle in proceedingsResearchpeer-review

Standard

Interaction is More Beneficial in Complex Reinforcement Learning Problems than in Simple Ones. / Stahlhut, Chris; Navarro-Guerrero, Nicolás; Weber, Cornelius; Wermter, Stefan.

Proceedings of the Interdisziplinärer Workshop Kognitive Systeme. Bielefeld, Germany, 2015. p. 142-150.

Research output: Contribution to book/anthology/report/proceedingArticle in proceedingsResearchpeer-review

Harvard

Stahlhut, C, Navarro-Guerrero, N, Weber, C & Wermter, S 2015, Interaction is More Beneficial in Complex Reinforcement Learning Problems than in Simple Ones. in Proceedings of the Interdisziplinärer Workshop Kognitive Systeme. Bielefeld, Germany, pp. 142-150.

APA

Stahlhut, C., Navarro-Guerrero, N., Weber, C., & Wermter, S. (2015). Interaction is More Beneficial in Complex Reinforcement Learning Problems than in Simple Ones. In Proceedings of the Interdisziplinärer Workshop Kognitive Systeme (pp. 142-150). Bielefeld, Germany.

CBE

Stahlhut C, Navarro-Guerrero N, Weber C, Wermter S. 2015. Interaction is More Beneficial in Complex Reinforcement Learning Problems than in Simple Ones. In Proceedings of the Interdisziplinärer Workshop Kognitive Systeme. Bielefeld, Germany. pp. 142-150.

MLA

Stahlhut, Chris et al. "Interaction is More Beneficial in Complex Reinforcement Learning Problems than in Simple Ones". Proceedings of the Interdisziplinärer Workshop Kognitive Systeme. Bielefeld, Germany. 2015, 142-150.

Vancouver

Stahlhut C, Navarro-Guerrero N, Weber C, Wermter S. Interaction is More Beneficial in Complex Reinforcement Learning Problems than in Simple Ones. In Proceedings of the Interdisziplinärer Workshop Kognitive Systeme. Bielefeld, Germany. 2015. p. 142-150

Author

Stahlhut, Chris ; Navarro-Guerrero, Nicolás ; Weber, Cornelius ; Wermter, Stefan. / Interaction is More Beneficial in Complex Reinforcement Learning Problems than in Simple Ones. Proceedings of the Interdisziplinärer Workshop Kognitive Systeme. Bielefeld, Germany, 2015. pp. 142-150

Bibtex

@inproceedings{266c9a6f735c42fba0c53cd41ce88d90,
title = "Interaction is More Beneficial in Complex Reinforcement Learning Problems than in Simple Ones",
abstract = "Giving interactive feedback, other than well done / badly done alone, can speed up reinforcement learning. However, the amount of feedback needed to improve the learning speed and performance has not been thoroughly investigated. To narrow this gap, we study the effects of one type of interaction: we allow the learner to ask a teacher whether the last performed action was good or not and if not, the learner can undo that action and choose another one; hence the learner avoids bad action sequences. This allows the interactive learner to reduce the overall number of steps necessary to reach its goal and learn faster than a non-interactive learner. Our results show that while interaction does not increase the learning speed in a simple task with 1 degree of freedom, it does speed up learning significantly in more complex tasks with 2 or 3 degrees of freedom.",
keywords = "ausrl",
author = "Chris Stahlhut and Nicol{\'a}s Navarro-Guerrero and Cornelius Weber and Stefan Wermter",
year = "2015",
month = "3",
day = "1",
language = "Udefineret/Ukendt",
pages = "142--150",
booktitle = "Proceedings of the Interdisziplin{\"a}rer Workshop Kognitive Systeme",

}

RIS

TY - GEN

T1 - Interaction is More Beneficial in Complex Reinforcement Learning Problems than in Simple Ones

AU - Stahlhut, Chris

AU - Navarro-Guerrero, Nicolás

AU - Weber, Cornelius

AU - Wermter, Stefan

PY - 2015/3/1

Y1 - 2015/3/1

N2 - Giving interactive feedback, other than well done / badly done alone, can speed up reinforcement learning. However, the amount of feedback needed to improve the learning speed and performance has not been thoroughly investigated. To narrow this gap, we study the effects of one type of interaction: we allow the learner to ask a teacher whether the last performed action was good or not and if not, the learner can undo that action and choose another one; hence the learner avoids bad action sequences. This allows the interactive learner to reduce the overall number of steps necessary to reach its goal and learn faster than a non-interactive learner. Our results show that while interaction does not increase the learning speed in a simple task with 1 degree of freedom, it does speed up learning significantly in more complex tasks with 2 or 3 degrees of freedom.

AB - Giving interactive feedback, other than well done / badly done alone, can speed up reinforcement learning. However, the amount of feedback needed to improve the learning speed and performance has not been thoroughly investigated. To narrow this gap, we study the effects of one type of interaction: we allow the learner to ask a teacher whether the last performed action was good or not and if not, the learner can undo that action and choose another one; hence the learner avoids bad action sequences. This allows the interactive learner to reduce the overall number of steps necessary to reach its goal and learn faster than a non-interactive learner. Our results show that while interaction does not increase the learning speed in a simple task with 1 degree of freedom, it does speed up learning significantly in more complex tasks with 2 or 3 degrees of freedom.

KW - ausrl

M3 - Konferencebidrag i proceedings

SP - 142

EP - 150

BT - Proceedings of the Interdisziplinärer Workshop Kognitive Systeme

CY - Bielefeld, Germany

ER -