Decoupling feasibility from objective optimization by training the RL policy over latent actions that map to feasible actions improves sample efficiency and constraint satisfaction in continuous constrained RL.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2024 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Action Mapping for Reinforcement Learning in Continuous Environments with Constraints
Decoupling feasibility from objective optimization by training the RL policy over latent actions that map to feasible actions improves sample efficiency and constraint satisfaction in continuous constrained RL.