{"id":31021447,"url":"https://github.com/sisl/madrl","last_synced_at":"2025-09-13T11:21:46.157Z","repository":{"id":44163842,"uuid":"62418890","full_name":"sisl/MADRL","owner":"sisl","description":"Repo containing code for multi-agent deep reinforcement learning (MADRL). ","archived":false,"fork":false,"pushed_at":"2023-04-12T16:04:09.000Z","size":453,"stargazers_count":663,"open_issues_count":8,"forks_count":123,"subscribers_count":20,"default_branch":"master","last_synced_at":"2024-10-31T07:42:56.613Z","etag":null,"topics":[],"latest_commit_sha":null,"homepage":null,"language":"Python","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"mit","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/sisl.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":"LICENSE","code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null}},"created_at":"2016-07-01T20:47:43.000Z","updated_at":"2024-10-30T07:12:02.000Z","dependencies_parsed_at":"2022-09-09T08:51:19.164Z","dependency_job_id":null,"html_url":"https://github.com/sisl/MADRL","commit_stats":null,"previous_names":[],"tags_count":1,"template":false,"template_full_name":null,"purl":"pkg:github/sisl/MADRL","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/sisl%2FMADRL","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/sisl%2FMADRL/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/sisl%2FMADRL/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/sisl%2FMADRL/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/sisl","download_url":"https://codeload.github.com/sisl/MADRL/tar.gz/refs/heads/master","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/sisl%2FMADRL/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":274955841,"owners_count":25380669,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2022-07-04T15:15:14.044Z","status":"online","status_checked_at":"2025-09-13T02:00:10.085Z","response_time":70,"last_error":null,"robots_txt_status":"success","robots_txt_updated_at":"2025-07-24T06:49:26.215Z","robots_txt_url":"https://github.com/robots.txt","online":true,"can_crawl_api":true,"host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":[],"created_at":"2025-09-13T11:21:43.783Z","updated_at":"2025-09-13T11:21:46.142Z","avatar_url":"https://github.com/sisl.png","language":"Python","funding_links":[],"categories":[],"sub_categories":[],"readme":"# MADRL\n\n**Note**: \n**A maintained version of the first three environments with various fixes is included with PettingZoo (https://github.com/PettingZoo-Team/PettingZoo, https://pettingzoo.farama.org/environments/sisl/)**\n\nThis package provides implementations of the following multi-agent reinforcement learning environemnts:\n\n- [Pursuit Evastion](https://github.com/sisl/MADRL/blob/master/madrl_environments/pursuit/pursuit_evade.py)\n- [Waterworld](https://github.com/sisl/MADRL/blob/master/madrl_environments/pursuit/waterworld.py)\n- [Multi-Agent Walker](https://github.com/sisl/MADRL/blob/master/madrl_environments/walker/multi_walker.py)\n- [Multi-Ant](https://github.com/sisl/MADRL/blob/master/madrl_environments/mujoco/ant/multi_ant.py)\n\n\n## Requirements\n\nThis package requires both [OpenAI Gym](https://github.com/openai/gym) and a forked version of [rllab](https://github.com/rejuvyesh/rllab/tree/multiagent) (the multiagent branch). There are a number of other requirements which can be found\nin `rllab/environment.yml` file if using `anaconda` distribution.\n\n## Setup\n\nThe easiest way to install MADRL and its dependencies is to perform a recursive clone of this repository.\n```bash\ngit clone --recursive git@github.com:sisl/MADRL.git\n```\n\nThen, add directories to `PYTHONPATH`\n```bash\nexport PYTHONPATH=$(pwd):$(pwd)/rltools:$(pwd)/rllab:$PYTHONPATH\n```\n\nInstall the required dependencies. Good idea is to look into `rllab/environment.yml` file if using `anaconda` distribution.\n\n## Usage\n\nExample run with curriculum:\n\n```bash\npython3 runners/run_multiwalker.py rllab \\ # Use rllab for training\n    --control decentralized \\ # Decentralized training protocol\n    --policy_hidden 100,50,25 \\ # Set MLP policy hidden layer sizes\n    --n_iter 200 \\ # Number of iterations\n    --n_walkers 2 \\ # Starting number of walkers\n    --batch_size 24000 \\ # Number of rollout waypoints\n    --curriculum lessons/multiwalker/env.yaml\n```\n\n## Details\n\nPolicy definitions exist in `rllab/sandbox/rocky/tf/policies`.\n\n## Citation\n\nPlease cite the accompanied paper, if you find this useful:\n\n```\n@inproceedings{gupta2017cooperative,\n  title={Cooperative multi-agent control using deep reinforcement learning},\n  author={Gupta, Jayesh K and Egorov, Maxim and Kochenderfer, Mykel},\n  booktitle={International Conference on Autonomous Agents and Multiagent Systems},\n  pages={66--83},\n  year={2017},\n  organization={Springer}\n}\n```\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fsisl%2Fmadrl","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fsisl%2Fmadrl","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fsisl%2Fmadrl/lists"}