Guaranteed hierarchical reinforcement learning
Reinforcement learning (RL) is a sub-field of machine learning that aims to train an agent in an interactive environment to sequentially make choices via a process of trial-and-error, to maximize a total reward over time. RL has been studied for decades and has a strong and established theoretica...
| Main Author: | |
|---|---|
| Other Authors: | |
| Format: | Final Year Project (FYP) |
| Language: | English |
| Published: |
Nanyang Technological University
2024
|
| Subjects: | |
| Online Access: | https://hdl.handle.net/10356/175473 |
| _version_ | 1811697190192545792 |
|---|---|
| author | Ang, Riley Xile |
| author2 | Arvind Easwaran |
| author_facet | Arvind Easwaran Ang, Riley Xile |
| author_sort | Ang, Riley Xile |
| collection | NTU |
| description | Reinforcement learning (RL) is a sub-field of machine learning that aims to train an agent in an interactive environment to sequentially make choices via a process of trial-and-error, to maximize a total reward over time.
RL has been studied for decades and has a strong and established theoretical foundation. Practically, it has gained prominence owing to projects in a wide range of fields including gaming, robotics, automation, etc. Despite its contributions and rise to popularity, RL is often resource-intensive in both its training time and memory requirements. Successfully training an agent with low margin of errors and high confidence bounds continues to remain an open research problem.
Consequently, the focus of this project will be to use existing RL algorithms, particularly Speedy Q-Learning (SQL), a variant of tabular model-free Q-Learning, to design a Hierarchical Reinforcement Learning (HRL) agent in a continuous state space setting. Additionally, this project aims to evaluate the overall performance of the agent against proven theoretical bounds with the Probably Approximately Correct (PAC) framework. |
| first_indexed | 2024-10-01T07:51:19Z |
| format | Final Year Project (FYP) |
| id | ntu-10356/175473 |
| institution | Nanyang Technological University |
| language | English |
| last_indexed | 2024-10-01T07:51:19Z |
| publishDate | 2024 |
| publisher | Nanyang Technological University |
| record_format | dspace |
| spelling | ntu-10356/1754732024-04-26T15:45:19Z Guaranteed hierarchical reinforcement learning Ang, Riley Xile Arvind Easwaran School of Computer Science and Engineering arvinde@ntu.edu.sg Computer and Information Science Reinforcement learning PAC Hierarchical reinforcement learning Q-learning Reinforcement learning (RL) is a sub-field of machine learning that aims to train an agent in an interactive environment to sequentially make choices via a process of trial-and-error, to maximize a total reward over time. RL has been studied for decades and has a strong and established theoretical foundation. Practically, it has gained prominence owing to projects in a wide range of fields including gaming, robotics, automation, etc. Despite its contributions and rise to popularity, RL is often resource-intensive in both its training time and memory requirements. Successfully training an agent with low margin of errors and high confidence bounds continues to remain an open research problem. Consequently, the focus of this project will be to use existing RL algorithms, particularly Speedy Q-Learning (SQL), a variant of tabular model-free Q-Learning, to design a Hierarchical Reinforcement Learning (HRL) agent in a continuous state space setting. Additionally, this project aims to evaluate the overall performance of the agent against proven theoretical bounds with the Probably Approximately Correct (PAC) framework. Bachelor's degree 2024-04-24T13:19:23Z 2024-04-24T13:19:23Z 2024 Final Year Project (FYP) Ang, R. X. (2024). Guaranteed hierarchical reinforcement learning. Final Year Project (FYP), Nanyang Technological University, Singapore. https://hdl.handle.net/10356/175473 https://hdl.handle.net/10356/175473 en application/pdf Nanyang Technological University |
| spellingShingle | Computer and Information Science Reinforcement learning PAC Hierarchical reinforcement learning Q-learning Ang, Riley Xile Guaranteed hierarchical reinforcement learning |
| title | Guaranteed hierarchical reinforcement learning |
| title_full | Guaranteed hierarchical reinforcement learning |
| title_fullStr | Guaranteed hierarchical reinforcement learning |
| title_full_unstemmed | Guaranteed hierarchical reinforcement learning |
| title_short | Guaranteed hierarchical reinforcement learning |
| title_sort | guaranteed hierarchical reinforcement learning |
| topic | Computer and Information Science Reinforcement learning PAC Hierarchical reinforcement learning Q-learning |
| url | https://hdl.handle.net/10356/175473 |
| work_keys_str_mv | AT angrileyxile guaranteedhierarchicalreinforcementlearning |