A toy implementation of paper "Meta-Gradient Reinforcement Learning"
Xu, Zhongwen, Hado P. van Hasselt, and David Silver. "Meta-gradient reinforcement learning." Advances in neural information processing systems 31 (2018). Link
In the very beginning, we start with A2C under CartPole-v1 environment and examine the property of Meta-gradient Learning(MGL) on the discount factor γ.
- Pytorch
- OpenAI Gym
p.s.: For now all the scripts are CPU versions.