COLA

Note

COLA is the first work to address the Dec-POMDP problem by integrating contrastive learning into multi-agent reinforcement learning. And we believe the COLA framework is the most cost-effective, bringing remarkable performance improvement with minor changes of reinforcement learning models. Our approach of simply adding a one-hot consensus encoding to the network input can be extended to any other multi-agent reinforcement learning algorithm.

The implementation of the following methods can also be found in this codebase, which are finished by the authors of PyMARL. PyMARL is written in PyTorch and uses SMAC as its environment.

Installation instructions

Build the Dockerfile using

cd docker
bash build.sh

Set up StarCraft II and SMAC:

bash install_sc2.sh

This will download SC2 into the 3rdparty folder and copy the maps necessary to run over.

It is worth noting that we run the all experiments on SC2.4.6.2.69232, not easier SC2.4.10. Performance is not always comparable between versions.

The requirements.txt file can be used to install the necessary packages into a virtual environment (not recomended).

Set up Google Research Football: Please follow the Quick Start in https://github.com/google-research/football.

Run an experiment

python3 src/main.py --config=cola --env-config=sc2 with env_args.map_name=2s3z env_args.seed=1

python3 src/main.py --config=cola --env-config=academy_3_vs_1_with_keeper with seed=1

The config files act as defaults for an algorithm or environment.

They are all located in src/config. --config refers to the config files in src/config/algs --env-config refers to the config files in src/config/envs

All results will be stored in the Results folder.

The previous config files used for the SMAC Beta have the suffix _beta.

Saving and loading learnt models

Saving models

You can save the learnt models to disk by setting save_model = True, which is set to False by default. The frequency of saving models can be adjusted using save_model_interval configuration. Models will be saved in the result directory, under the folder called models. The directory corresponding each run will contain models saved throughout the experiment, each within a folder corresponding to the number of timesteps passed since starting the learning process.

Loading models

Learnt models can be loaded using the checkpoint_path parameter, after which the learning will proceed from the corresponding timestep.

Watching StarCraft II replays

save_replay option allows saving replays of models which are loaded using checkpoint_path. Once the model is successfully loaded, test_nepisode number of episodes are run on the test mode and a .SC2Replay file is saved in the Replay directory of StarCraft II. Please make sure to use the episode runner if you wish to save a replay, i.e., runner=episode. The name of the saved replay file starts with the given env_args.save_replay_prefix (map_name if empty), followed by the current timestamp.

The saved replays can be watched by double-clicking on them or using the following command:

python -m pysc2.bin.play --norender --rgb_minimap_size 0 --replay NAME.SC2Replay

Note: Replays cannot be watched using the Linux version of StarCraft II. Please use either the Mac or Windows version of the StarCraft II client.

deligentfool/COLA