Evaluation of Reinforcement Learning Methods for a Self-learning System

David Bechtold, Alexander Wendt, Axel Jantsch

2020

Abstract

In recent years, interest in self-learning methods has increased significantly. A challenge is to learn to survive in a real or simulated world by solving tasks with as little prior knowledge about itself, the task, and the environment. In this paper, the state of the art methods of reinforcement learning, in particular, Q-learning, are analyzed regarding applicability to such a problem. The Q-learning algorithm is completed with replay memories and exploration functions. Several small improvements are proposed. The methods are then evaluated in two simulated environments: a discrete bit-flip and a continuous pendulum environment. The result is a lookup table of the best suitable algorithms for each type of problem.

Paper Citation

in Harvard Style

Bechtold D., Wendt A. and Jantsch A. (2020). Evaluation of Reinforcement Learning Methods for a Self-learning System. In Proceedings of the 12th International Conference on Agents and Artificial Intelligence - Volume 2: ICAART, ISBN 978-989-758-395-7, pages 36-47. DOI: 10.5220/0008909500360047

in Bibtex Style

@conference{icaart20,
author={David Bechtold and Alexander Wendt and Axel Jantsch},
title={Evaluation of Reinforcement Learning Methods for a Self-learning System},
booktitle={Proceedings of the 12th International Conference on Agents and Artificial Intelligence - Volume 2: ICAART,},
year={2020},
pages={36-47},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0008909500360047},
isbn={978-989-758-395-7},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 12th International Conference on Agents and Artificial Intelligence - Volume 2: ICAART,
TI - Evaluation of Reinforcement Learning Methods for a Self-learning System
SN - 978-989-758-395-7
AU - Bechtold D.
AU - Wendt A.
AU - Jantsch A.
PY - 2020
SP - 36
EP - 47
DO - 10.5220/0008909500360047