Scalable Asynchronous Federated Deep Reinforcement Learning Simulator
AFDRL simulates a large-scale deterministic federation of clients seeking to maximize reward in an ALE game.
CMake is required to build afdrl.
$ mkdir build
$ cd build
$ cmake -DCMAKE_BUILD_TYPE=Release ..
$ make
$ afdrl/afdrlTo run a simulation of 64 participants, execute the following:
$ afdrl/afdrl -c 64 --gpu