TY - RPRT TI - Model-free Reinforcement Learning for Model-based Control: Towards Safe, Interpretable and Sample-efficient Agents AU - Thomas Banker AU - Ali Mesbah PY - 2025 UR - https://arxiv.org/abs/2507.13491 ID - 2507.13491 ER -