نسخة أولية وصول مفتوح
CORE-RL: Confidence-Oriented Reliability Evaluation of Black-Box Reinforcement Learning Policies
The deployment of Reinforcement Learning (RL) agents in critical domains must be preceded with a pipeline to evaluate the alignment of the RL agent with complex multi-objective specifications and robustness under real-world environmental drift. However, to protect intellectual property, the RL agent may be delivered fo …