Towards Interpretable Reinforcement Learning with Constrained Normalizing Flow Policies
- Finn Rietz,
- Erik Schaffernicht,
- ,
- Johannes A. Stork
- Örebro University,
- ,
- ,
Research Output:
Conference Article in Proceeding or Book/Report chapter
Article in proceedings
Peer-reviewOpen access
Publication Information
Output type
Research Output:
Conference Article in Proceeding or Book/Report chapter
Article in proceedings
Peer-reviewOriginal language
EnglishPublication milestones
- Published - 17/05/2024
Publication status
Published - 17/05/2024
Place of publication
Vienna, ATHost publication title
Proceedings of the ICRA2024 Workshop on Human-aligned Reinforcement Learning for Autonomous Agents and RobotsAbstract
Reinforcement learning policies are typically represented by black-box neural networks, which are non-interpretable and not well-suited for safety-critical domains. To address both of these issues, we propose constrained normalizing flow policies as interpretable and safe-by-construction policy models. We achieve safety for reinforcement learning problems with instantaneous safety constraints, for which we can exploit domain knowledge by analytically constructing a normalizing flow that ensures constraint satisfaction. The normalizing flow corresponds to an interpretable sequence of transformations on action samples, each ensuring alignment with respect to a particular constraint. Our experiments reveal benefits beyond interpretability in an easier learning objective and maintained constraint satisfaction throughout the entire learning process. Our approach leverages constraints over reward engineering while offering enhanced interpretability, safety, and direct means of providing domain knowledge to the agent without relying on complex reward functions.
Access to documents
License:Unspecified
Final published version
License:Unspecified
Related Event
Title
Workshop on Human-aligned Reinforcement Learning for Autonomous Agents and Robots
Event type
WorkshopDegree of recognition
International eventDate
17/05/2024 Location
YokohamaJapan
