PID Accelerated Temporal Difference Algorithms
Fuente:
arXiv
Guardado en:
| Autores principales: | , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2024
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866914933688500224 |
|---|---|
| author | Bedaywi, Mark Rakhsha, Amin Farahmand, Amir-massoud |
| author_facet | Bedaywi, Mark Rakhsha, Amin Farahmand, Amir-massoud |
| contents | Long-horizon tasks, which have a large discount factor, pose a challenge for most conventional reinforcement learning (RL) algorithms. Algorithms such as Value Iteration and Temporal Difference (TD) learning have a slow convergence rate and become inefficient in these tasks. When the transition distributions are given, PID VI was recently introduced to accelerate the convergence of Value Iteration using ideas from control theory. Inspired by this, we introduce PID TD Learning and PID Q-Learning algorithms for the RL setting, in which only samples from the environment are available. We give a theoretical analysis of the convergence of PID TD Learning and its acceleration compared to the conventional TD Learning. We also introduce a method for adapting PID gains in the presence of noise and empirically verify its effectiveness. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2407_08803 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | PID Accelerated Temporal Difference Algorithms Bedaywi, Mark Rakhsha, Amin Farahmand, Amir-massoud Machine Learning Artificial Intelligence Systems and Control Optimization and Control Long-horizon tasks, which have a large discount factor, pose a challenge for most conventional reinforcement learning (RL) algorithms. Algorithms such as Value Iteration and Temporal Difference (TD) learning have a slow convergence rate and become inefficient in these tasks. When the transition distributions are given, PID VI was recently introduced to accelerate the convergence of Value Iteration using ideas from control theory. Inspired by this, we introduce PID TD Learning and PID Q-Learning algorithms for the RL setting, in which only samples from the environment are available. We give a theoretical analysis of the convergence of PID TD Learning and its acceleration compared to the conventional TD Learning. We also introduce a method for adapting PID gains in the presence of noise and empirically verify its effectiveness. |
| title | PID Accelerated Temporal Difference Algorithms |
| topic | Machine Learning Artificial Intelligence Systems and Control Optimization and Control |
| url | https://arxiv.org/abs/2407.08803 |