Noise Parameterization of Continuous Deep Reinforcement Learning for a Class of Non-linear System
Reinforcement learning (RL) is one of the most important algorithms for artificial intelligence. DDPG as continuous controller approach which can work at continuous and high dimensional data applies in this paper to solve nonlinear valve system. The aim of this paper is gaining analysis of the DDPG...
| Main Authors: | , , |
|---|---|
| Format: | Conference or Workshop Item |
| Language: | English |
| Published: |
2022
|
| Subjects: | |
| Online Access: | https://repository.ugm.ac.id/282134/1/Surriani%20et%20al%20-%202022%20-%20Noise_Parameterization_of_Continuous_Deep_Reinforcement_Learning_for_a_Class_of_Non-linear_System.pdf |
| _version_ | 1826050492977905664 |
|---|---|
| author | Surriani, Atikah Wahyunggoro, Oyas Cahyadi, Adha Imam |
| author_facet | Surriani, Atikah Wahyunggoro, Oyas Cahyadi, Adha Imam |
| author_sort | Surriani, Atikah |
| collection | UGM |
| description | Reinforcement learning (RL) is one of the most important algorithms for artificial intelligence. DDPG as continuous controller approach which can work at continuous and high dimensional data applies in this paper to solve nonlinear valve system. The aim of this paper is gaining analysis of the DDPG noise parameterization. Noise parameter addition is known to be able to increase the exploration ability of the algorithm. The noise parameterization is using the Ornstein-Uhlenbeck (OU) noise injection. This exploration investigation concerns to the algorithm's performance. The evaluation measurement is based on the total reward to system during training. The result indicates that noise parameterization affects the performance of the algorithm. The comparisons show that the injection of OU noise for DDPG algorithm influences the total reward. The simulation find that the total reward that is achieved by DDPG with OU noise injection is higher than DDPG without OU noise injection at 317,810. © 2022 IEEE. |
| first_indexed | 2024-03-14T00:04:51Z |
| format | Conference or Workshop Item |
| id | oai:generic.eprints.org:282134 |
| institution | Universiti Gadjah Mada |
| language | English |
| last_indexed | 2024-03-14T00:04:51Z |
| publishDate | 2022 |
| record_format | dspace |
| spelling | oai:generic.eprints.org:2821342023-11-29T08:36:57Z https://repository.ugm.ac.id/282134/ Noise Parameterization of Continuous Deep Reinforcement Learning for a Class of Non-linear System Surriani, Atikah Wahyunggoro, Oyas Cahyadi, Adha Imam Electrical and Electronic Engineering not elsewhere classified Reinforcement learning (RL) is one of the most important algorithms for artificial intelligence. DDPG as continuous controller approach which can work at continuous and high dimensional data applies in this paper to solve nonlinear valve system. The aim of this paper is gaining analysis of the DDPG noise parameterization. Noise parameter addition is known to be able to increase the exploration ability of the algorithm. The noise parameterization is using the Ornstein-Uhlenbeck (OU) noise injection. This exploration investigation concerns to the algorithm's performance. The evaluation measurement is based on the total reward to system during training. The result indicates that noise parameterization affects the performance of the algorithm. The comparisons show that the injection of OU noise for DDPG algorithm influences the total reward. The simulation find that the total reward that is achieved by DDPG with OU noise injection is higher than DDPG without OU noise injection at 317,810. © 2022 IEEE. 2022 Conference or Workshop Item PeerReviewed application/pdf en https://repository.ugm.ac.id/282134/1/Surriani%20et%20al%20-%202022%20-%20Noise_Parameterization_of_Continuous_Deep_Reinforcement_Learning_for_a_Class_of_Non-linear_System.pdf Surriani, Atikah and Wahyunggoro, Oyas and Cahyadi, Adha Imam (2022) Noise Parameterization of Continuous Deep Reinforcement Learning for a Class of Non-linear System. In: 2022 14th International Conference on Information Technology and Electrical Engineering (ICITEE). https://ieeexplore.ieee.org/document/9954121 |
| spellingShingle | Electrical and Electronic Engineering not elsewhere classified Surriani, Atikah Wahyunggoro, Oyas Cahyadi, Adha Imam Noise Parameterization of Continuous Deep Reinforcement Learning for a Class of Non-linear System |
| title | Noise Parameterization of Continuous Deep Reinforcement Learning for a Class of Non-linear System |
| title_full | Noise Parameterization of Continuous Deep Reinforcement Learning for a Class of Non-linear System |
| title_fullStr | Noise Parameterization of Continuous Deep Reinforcement Learning for a Class of Non-linear System |
| title_full_unstemmed | Noise Parameterization of Continuous Deep Reinforcement Learning for a Class of Non-linear System |
| title_short | Noise Parameterization of Continuous Deep Reinforcement Learning for a Class of Non-linear System |
| title_sort | noise parameterization of continuous deep reinforcement learning for a class of non linear system |
| topic | Electrical and Electronic Engineering not elsewhere classified |
| url | https://repository.ugm.ac.id/282134/1/Surriani%20et%20al%20-%202022%20-%20Noise_Parameterization_of_Continuous_Deep_Reinforcement_Learning_for_a_Class_of_Non-linear_System.pdf |
| work_keys_str_mv | AT surrianiatikah noiseparameterizationofcontinuousdeepreinforcementlearningforaclassofnonlinearsystem AT wahyunggorooyas noiseparameterizationofcontinuousdeepreinforcementlearningforaclassofnonlinearsystem AT cahyadiadhaimam noiseparameterizationofcontinuousdeepreinforcementlearningforaclassofnonlinearsystem |