Comparison and analysis of DQN performance with different hyperparameters

Research Article
Open access

Comparison and analysis of DQN performance with different hyperparameters

Zhuoxian Huang 1* , Jiayi Ou 2 , Ming Wang 3
  • 1 University of Birmingham    
  • 2 Nanjing University of Aeronautics and Astronautics    
  • 3 Hebei University of Technology    
  • *corresponding author zxh131@student.bham.ac.uk
Published on 23 October 2023 | https://doi.org/10.54254/2755-2721/15/20230806
ACE Vol.15
ISSN (Print): 2755-273X
ISSN (Online): 2755-2721
ISBN (Print): 978-1-83558-021-9​
ISBN (Online): 978-1-83558-022-6

Abstract

Deep Q-learning Network (DQN) is an algorithm that combines Q-learning and deep neural network, its model can adopt high-dimensional input and low-dimensional output. As a deep reinforcement learning algorithm proposed ten years ago, its performance on some Atari games has surpassed all previous algorithms, even some human experts, which fully reflects DQN’s high research value. The tuning of hyperparameters is crucial for any algorithm, especially for those with strong performance. The same algorithm can produce completely different results when using different sets of hyperparameters, and suitable values can considerably improve the algorithm. Based on the DQN we implement, we test on number of episodes, size of replay buffer, gamma, learning rate and batch size with different values. In each round of experiments, except for the target hyperparameter, all others use default values, and we recorded the impact of these changes on training performance. The result indicates that as the number of episodes continues to increase, the performance improves steadily and degressively. The same conclusion is also applicable to the size of replay buffer, while other hyperparameters need to be given values to have optimal performance.

Keywords:

DQN, performance, hyperparameters, comparison

Huang,Z.;Ou,J.;Wang,M. (2023). Comparison and analysis of DQN performance with different hyperparameters. Applied and Computational Engineering,15,22-29.
Export citation

References

[1]. Hoffer, E., Hubara, I., & Soudry, D. Train longer, generalize better: closing the generalization gap in large batch training of neural networks. 2017 ArXiv (Cornell University). https://arxiv.org/pdf/1705.08741.pdf

[2]. Mnih, V., Kavukcuoglu, K., Silver, D., Graves, A., Antonoglou, I., Wierstra, D., & Riedmiller, M. Playing Atari with Deep Reinforcement Learning. 2013 Cornell University. http://cs.nyu.edu/~koray/publis/mnih-atari-2013.pdf

[3]. Mnih, V., Kavukcuoglu, K., Silver, D., Hassabis, D. Human-level control through deep reinforcement learning. 2015 Nature, 518(7540), 529–533.

[4]. Reddi, S. J., Kale, S., & Kumar, S. On the Convergence of Adam and Beyond. 2018 International Conference on Learning Representations 38(12) 233-244.

[5]. Smith, S. G., Kindermans, P., Ying, C., & Le, Q. V. Don’t decay the learning rate, increase the batch size. 2018 International Conference on Learning Representations. 312-324.

[6]. Wilson, A. C., Roelofs, R., Stern, M., Srebro, N., & Recht, B. The Marginal Value of Adaptive Gradient Methods in Machine Learning. 2017, Neural Information Processing Systems, 30, 4148–4158.

[7]. You, K., Long, M., Wang, J., & Jordan, M. I. How Does Learning Rate Decay Help Modern Neural Networks. 2019, ArXiv Cornell University. https://arxiv.org/pdf/1908.01878.pdf

[8]. Zhang, S., & Sutton, R.. A Deeper Look at Experience Replay. 2017 Cornell University. https://arxiv.org/pdf/1712.01275.pdf

[9]. Li H , Kumar N , Chen R , et al. Deep Reinforcement Learning. 2018 IEEE International Conference on Acoustics, Speech and Signal Processing,349-359.

[10]. Lillicrap T P, Hunt J J, Pritzel A et al. Continuous control with deep reinforcement learning. 2015 Computer science, 101-115.

Cite this article

Huang,Z.;Ou,J.;Wang,M. (2023). Comparison and analysis of DQN performance with different hyperparameters. Applied and Computational Engineering,15,22-29.

Data availability

The datasets used and/or analyzed during the current study will be available from the authors upon reasonable request.

Disclaimer/Publisher's Note

The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of EWA Publishing and/or the editor(s). EWA Publishing and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

About volume

Volume title: Proceedings of the 5th International Conference on Computing and Data Science

ISBN：978-1-83558-021-9​(Print) / 978-1-83558-022-6(Online)
Editor：Marwan Omar, Roman Bauer, Alan Wang
Conference website: https://2023.confcds.org/
Conference date: 14 July 2023
Series: Applied and Computational Engineering
Volume number: Vol.15
ISSN：2755-2721(Print) / 2755-273X(Online)

© 2024 by the author(s). Licensee EWA Publishing, Oxford, UK. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license. Authors who publish this series agree to the following terms:
1. Authors retain copyright and grant the series right of first publication with the work simultaneously licensed under a Creative Commons Attribution License that allows others to share the work with an acknowledgment of the work's authorship and initial publication in this series.
2. Authors are able to enter into separate, additional contractual arrangements for the non-exclusive distribution of the series's published version of the work (e.g., post it to an institutional repository or publish it in a book), with an acknowledgment of its initial publication in this series.
3. Authors are permitted and encouraged to post their work online (e.g., in institutional repositories or on their website) prior to and during the submission process, as it can lead to productive exchanges, as well as earlier and greater citation of published work (See Open access policy for details).

References

[1]. Hoffer, E., Hubara, I., & Soudry, D. Train longer, generalize better: closing the generalization gap in large batch training of neural networks. 2017 ArXiv (Cornell University). https://arxiv.org/pdf/1705.08741.pdf

[2]. Mnih, V., Kavukcuoglu, K., Silver, D., Graves, A., Antonoglou, I., Wierstra, D., & Riedmiller, M. Playing Atari with Deep Reinforcement Learning. 2013 Cornell University. http://cs.nyu.edu/~koray/publis/mnih-atari-2013.pdf

[3]. Mnih, V., Kavukcuoglu, K., Silver, D., Hassabis, D. Human-level control through deep reinforcement learning. 2015 Nature, 518(7540), 529–533.

[4]. Reddi, S. J., Kale, S., & Kumar, S. On the Convergence of Adam and Beyond. 2018 International Conference on Learning Representations 38(12) 233-244.

[5]. Smith, S. G., Kindermans, P., Ying, C., & Le, Q. V. Don’t decay the learning rate, increase the batch size. 2018 International Conference on Learning Representations. 312-324.

[6]. Wilson, A. C., Roelofs, R., Stern, M., Srebro, N., & Recht, B. The Marginal Value of Adaptive Gradient Methods in Machine Learning. 2017, Neural Information Processing Systems, 30, 4148–4158.

[7]. You, K., Long, M., Wang, J., & Jordan, M. I. How Does Learning Rate Decay Help Modern Neural Networks. 2019, ArXiv Cornell University. https://arxiv.org/pdf/1908.01878.pdf

[8]. Zhang, S., & Sutton, R.. A Deeper Look at Experience Replay. 2017 Cornell University. https://arxiv.org/pdf/1712.01275.pdf

[9]. Li H , Kumar N , Chen R , et al. Deep Reinforcement Learning. 2018 IEEE International Conference on Acoustics, Speech and Signal Processing,349-359.

[10]. Lillicrap T P, Hunt J J, Pritzel A et al. Continuous control with deep reinforcement learning. 2015 Computer science, 101-115.

For authors
Guide for authors Policies for authors Aims and scope Subscription Open access policy
For editors
Policies for editors Editorial board
For reviewers
Policies for reviewers Privacy policy

Cookies are used by this site.
Copyright © 2024 EWA Publishing, its licensors, and contributors. All rights reserved, including text and data mining, AI training, and similar technologies.