Formal Ethical Obligations in Reinforcement Learning Agents: Verification and Policy Updates
Fuente:
arXiv
Saved in:
| Main Authors: | , |
|---|---|
| Format: | Preprint |
| Published: |
2024
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866929445200199680 |
|---|---|
| author | Shea-Blymyer, Colin Abbas, Houssam |
| author_facet | Shea-Blymyer, Colin Abbas, Houssam |
| contents | When designing agents for operation in uncertain environments, designers need tools to automatically reason about what agents ought to do, how that conflicts with what is actually happening, and how a policy might be modified to remove the conflict. These obligations include ethical and social obligations, permissions and prohibitions, which constrain how the agent achieves its mission and executes its policy. We propose a new deontic logic, Expected Act Utilitarian deontic logic, for enabling this reasoning at design time: for specifying and verifying the agent's strategic obligations, then modifying its policy from a reference policy to meet those obligations. Unlike approaches that work at the reward level, working at the logical level increases the transparency of the trade-offs. We introduce two algorithms: one for model-checking whether an RL agent has the right strategic obligations, and one for modifying a reference decision policy to make it meet obligations expressed in our logic. We illustrate our algorithms on DAC-MDPs which accurately abstract neural decision policies, and on toy gridworld environments. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2408_00147 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Formal Ethical Obligations in Reinforcement Learning Agents: Verification and Policy Updates Shea-Blymyer, Colin Abbas, Houssam Artificial Intelligence Logic in Computer Science When designing agents for operation in uncertain environments, designers need tools to automatically reason about what agents ought to do, how that conflicts with what is actually happening, and how a policy might be modified to remove the conflict. These obligations include ethical and social obligations, permissions and prohibitions, which constrain how the agent achieves its mission and executes its policy. We propose a new deontic logic, Expected Act Utilitarian deontic logic, for enabling this reasoning at design time: for specifying and verifying the agent's strategic obligations, then modifying its policy from a reference policy to meet those obligations. Unlike approaches that work at the reward level, working at the logical level increases the transparency of the trade-offs. We introduce two algorithms: one for model-checking whether an RL agent has the right strategic obligations, and one for modifying a reference decision policy to make it meet obligations expressed in our logic. We illustrate our algorithms on DAC-MDPs which accurately abstract neural decision policies, and on toy gridworld environments. |
| title | Formal Ethical Obligations in Reinforcement Learning Agents: Verification and Policy Updates |
| topic | Artificial Intelligence Logic in Computer Science |
| url | https://arxiv.org/abs/2408.00147 |