Comparison of Deep Q Network and Its Variations in a Banana Collecting Environment

DOI

10.2991/978-94-6463-370-2_18How to use a DOI?

Keywords

Reinforcement Learning; Deep Q-Network; Banans Collection

Abstract

Reinforcement Learning is widely applied in the field of virtual agent training, enabling them to accomplish specific tasks. The agent is trained to navigate and collect yellow bananas in a large, square world that contains many yellow bananas and blue bananas. The goal is to allow agents to collect as many yellow bananas and avoid as many blue bananas as possible within a limited number of training sessions, achieving a higher score. In this study, Deep Reinforcement Learning (DRL) algorithms are employed to train the agents. Three distinct methods, including Deep Q-Network (DQN), Double DQN (DDQN), and Dueling Double DQN (D3QN), are implemented in this project, and their performances are compared. It can be observed from their performance that, within three hundred time steps, the score rapidly ascends to approximately thirteen in the DQN algorithm, then starts to oscillate, while the mean value remains more stable in the DDQN algorithm and in the D3QN algorithm, the score increases at a relatively faster pace.

Copyright

© 2024 The Author(s)

Open Access

Open Access This chapter is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/), which permits any noncommercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

Download article (PDF)

TY  - CONF
AU  - Yifan Liu
PY  - 2024
DA  - 2024/02/14
TI  - Comparison of Deep Q Network and Its Variations in a Banana Collecting Environment
BT  - Proceedings of the 2023 International Conference on Data Science, Advanced Algorithm and Intelligent Computing (DAI 2023)
PB  - Atlantis Press
SP  - 156
EP  - 168
SN  - 1951-6851
UR  - https://doi.org/10.2991/978-94-6463-370-2_18
DO  - 10.2991/978-94-6463-370-2_18
ID  - Liu2024
ER  -