[homarp]

https://arxiv.org/abs/1611.02779 is the closest - RL2: Fast Reinforcement Learning via Slow Reinforcement Learning