A Pytorch implementation of The Predictron: End-To-End Learning and Planning, Silver et al.
- No Distributed GPU support
- Only tested for 1e5 time steps due to machine limits, and achieve a total MSE as 0.0035 only.
- The Predictron: End-To-End Learning and Planning, Silver et al. [link]
- brendanator/predictron
- zhongwen/predictron