Ablation Study of How Run Time Assurance Impacts the Training and Performance of Reinforcement Learning Agents
Fuente:
arXiv
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Preprint |
| Published: |
2022
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866913357260390400 |
|---|---|
| author | Hamilton, Nathaniel Dunlap, Kyle Johnson, Taylor T Hobbs, Kerianne L |
| author_facet | Hamilton, Nathaniel Dunlap, Kyle Johnson, Taylor T Hobbs, Kerianne L |
| contents | Reinforcement Learning (RL) has become an increasingly important research area as the success of machine learning algorithms and methods grows. To combat the safety concerns surrounding the freedom given to RL agents while training, there has been an increase in work concerning Safe Reinforcement Learning (SRL). However, these new and safe methods have been held to less scrutiny than their unsafe counterparts. For instance, comparisons among safe methods often lack fair evaluation across similar initial condition bounds and hyperparameter settings, use poor evaluation metrics, and cherry-pick the best training runs rather than averaging over multiple random seeds. In this work, we conduct an ablation study using evaluation best practices to investigate the impact of run time assurance (RTA), which monitors the system state and intervenes to assure safety, on effective learning. By studying multiple RTA approaches in both on-policy and off-policy RL algorithms, we seek to understand which RTA methods are most effective, whether the agents become dependent on the RTA, and the importance of reward shaping versus safe exploration in RL agent training. Our conclusions shed light on the most promising directions of SRL, and our evaluation methodology lays the groundwork for creating better comparisons in future SRL work. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2207_04117 |
| institution | arXiv |
| publishDate | 2022 |
| record_format | arxiv |
| spellingShingle | Ablation Study of How Run Time Assurance Impacts the Training and Performance of Reinforcement Learning Agents Hamilton, Nathaniel Dunlap, Kyle Johnson, Taylor T Hobbs, Kerianne L Machine Learning Systems and Control Reinforcement Learning (RL) has become an increasingly important research area as the success of machine learning algorithms and methods grows. To combat the safety concerns surrounding the freedom given to RL agents while training, there has been an increase in work concerning Safe Reinforcement Learning (SRL). However, these new and safe methods have been held to less scrutiny than their unsafe counterparts. For instance, comparisons among safe methods often lack fair evaluation across similar initial condition bounds and hyperparameter settings, use poor evaluation metrics, and cherry-pick the best training runs rather than averaging over multiple random seeds. In this work, we conduct an ablation study using evaluation best practices to investigate the impact of run time assurance (RTA), which monitors the system state and intervenes to assure safety, on effective learning. By studying multiple RTA approaches in both on-policy and off-policy RL algorithms, we seek to understand which RTA methods are most effective, whether the agents become dependent on the RTA, and the importance of reward shaping versus safe exploration in RL agent training. Our conclusions shed light on the most promising directions of SRL, and our evaluation methodology lays the groundwork for creating better comparisons in future SRL work. |
| title | Ablation Study of How Run Time Assurance Impacts the Training and Performance of Reinforcement Learning Agents |
| topic | Machine Learning Systems and Control |
| url | https://arxiv.org/abs/2207.04117 |