Comment by conjectures

Anyone have a reference relating the free energy minimisation principle / active inference to reinforcement learning type environments?