On the Effect of the Layers’ Number of Deep Neural Network for Improving the Reward of a Reinforcement Learning Robot

Talal Markabi; Bahaa Mansoura

doi:https://doi.org/10.54216/NIF.030202

Full Length Article

Volume 3 • Issue 2 • PP: 07–11 • 2024

On the Effect of the Layers’ Number of Deep Neural Network for Improving the Reward of a Reinforcement Learning Robot

Talal Markabi ^1*

mail

,

Bahaa Mansoura ¹

mail

¹Faculty of Informatics Engineering, Albaath University, Syria

* Corresponding Author.

DOI https://doi.org/10.54216/NIF.030202

format_quote Cite this article

verified

Open Access & Copyright

Received: November 26, 2023 Accepted: March 22, 2024

View PDF open_in_new

Abstract

The Q-learning algorithm in reinforcement learning is one of the algorithms that allows the robot to learn the surrounding environment without the need for prior training samples, using the principle of reward and punishment through interaction with the environment. Increasing the number of hidden layers of the deep neural network and adjusting some of its hyperparameters can increase the robot’s reward and thus obtain the best path to achieve the goal.

Keywords

Neural network Deep learning Robotics Layers

References

[1] V. Mnih, K. Kavukcuoglu, D. Silver, A. Graves,

I. Antonoglou, D. Wierstra, and M. Riedmiller, “Play-ing Atari with deep reinforcement learning,” NIPS Deep Learning Workshop, 2013.

[2] M. Naeem, S. T. H. Rizvi, and A. Coronato, “A gentle in-troduction to reinforcement learning and its application in different fields,” IEEE Access, vol. 8, pp. 209320–209344, 2020.

[3] S. Aradi, T. Becsi, and P. Gaspar, “Policy gradient based reinforcement learning approach for autonomous high-way driving,” in Proc. IEEE Conf. Control Technology and Applications, Copenhagen, Denmark, pp. 670–675, 2018.

[4] S. Doltsinis, P. Ferreira, and N. Lohse, “An MDP model-based reinforcement learning approach for pro-duction station ramp-up optimization: Q-learning anal-ysis,” IEEE Trans. Systems, Man, and Cybernetics: Sys-tems, vol. 44, no. 9, pp. 1125–1138, 2014.

[5] M. Ronecker and Y. Zhu, “Deep Q-network based deci-sion making for autonomous driving,” in Proc. 3rd Int. Conf. Robotics and Automation Sciences, pp. 154–160, 2019.

[6] A. Saini, T. Gupta, R. Kumar, A. Gupta, M. Panwar, and A. Mittal, “Image-based Indian monument recognition using convolutional neural networks,” in Proc. Int. Conf. Big Data, IoT and Data Science, pp. 138–142, 2017.

[7] E. Messner, M. Zohrer, and F. Pernkopf, “Heart sound segmentation—an event detection approach using deep recurrent neural networks,” IEEE Trans. Biomedical Engineering, vol. 65, no. 9, pp. 1964–1974, 2018.

[8] B. O’Donoghue, I. Osband, R. Munos, and V. Mnih, “The uncertainty Bellman equation and exploration,” in Proc. 35th Int. Conf. Machine Learning, vol. 80, pp. 3839–3848, 2018.

[9] A. Ng, Machine Learning. Stanford University, 2017.

[10] R. Rao and K. Narasimhan, “M-stage epsilon-greedy ex-ploration for reinforcement learning,” in AAAI-21 Work-shop on Reinforcement Learning in Games, 2021.

Cite This Article

Choose your preferred format

format_quote

Markabi, Talal, Mansoura, Bahaa. "On the Effect of the Layers’ Number of Deep Neural Network for Improving the Reward of a Reinforcement Learning Robot." Neutrosophic and Information Fusion, vol. 3, no. 2, 2024, pp. 07–11. DOI: https://doi.org/10.54216/NIF.030202

Markabi, T., Mansoura, B. (2024). On the Effect of the Layers’ Number of Deep Neural Network for Improving the Reward of a Reinforcement Learning Robot. Neutrosophic and Information Fusion, Volume 3(Issue 2), 07–11. DOI: https://doi.org/10.54216/NIF.030202

Markabi, Talal, Mansoura, Bahaa. "On the Effect of the Layers’ Number of Deep Neural Network for Improving the Reward of a Reinforcement Learning Robot." Neutrosophic and Information Fusion Volume 3, no. Issue 2 (2024): 07–11. DOI: https://doi.org/10.54216/NIF.030202

Markabi, T., Mansoura, B. (2024) 'On the Effect of the Layers’ Number of Deep Neural Network for Improving the Reward of a Reinforcement Learning Robot', Neutrosophic and Information Fusion, Volume 3(Issue 2), pp. 07–11. DOI: https://doi.org/10.54216/NIF.030202

Markabi T, Mansoura B. On the Effect of the Layers’ Number of Deep Neural Network for Improving the Reward of a Reinforcement Learning Robot. Neutrosophic and Information Fusion. 2024;Volume 3(Issue 2):07–11. DOI: https://doi.org/10.54216/NIF.030202

T. Markabi, B. Mansoura, "On the Effect of the Layers’ Number of Deep Neural Network for Improving the Reward of a Reinforcement Learning Robot," Neutrosophic and Information Fusion, vol. Volume 3, no. Issue 2, pp. 07–11, 2024. DOI: https://doi.org/10.54216/NIF.030202

policy

Publisher's Note

The statements, opinions, and data presented in this article are solely those of the author(s) and do not necessarily represent those of ASPG, the journal, or its editors. ASPG and the editors disclaim responsibility for any harm arising from the use of any ideas, methods, instructions, or products described in this article, to the fullest extent permitted by applicable law.

Digital Archive Ready