Let's Play Hanabi!
This repo contains the following algorithms to play Hanabi:
- Distributed Prioritized Experience Replay (need improvement)
- Bayesian Action Decoder (TODO: reorganize jupyter notebook)
In the future, all the implementations should go to https://github.com/JuliaReinforcementLearning/ReinforcementLearning.jl and the patch of Hanabi.jl in
src/hanabi_environment.jl should go to https://github.com/JuliaReinforcementLearning/ReinforcementLearningEnvironments.jl.