{"id":28549504,"url":"https://github.com/kempnerinstitute/nicewebrl","last_synced_at":"2025-07-14T17:37:53.537Z","repository":{"id":251638194,"uuid":"837994162","full_name":"KempnerInstitute/nicewebrl","owner":"KempnerInstitute","description":"NiceWebRL is a Python library for quickly making human subject experiments that leverage machine reinforcement learning environments.","archived":false,"fork":false,"pushed_at":"2025-07-13T18:48:11.000Z","size":633739,"stargazers_count":47,"open_issues_count":5,"forks_count":7,"subscribers_count":1,"default_branch":"main","last_synced_at":"2025-07-13T20:42:59.139Z","etag":null,"topics":[],"latest_commit_sha":null,"homepage":"","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/KempnerInstitute.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,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null,"zenodo":null}},"created_at":"2024-08-04T16:30:59.000Z","updated_at":"2025-07-12T17:12:49.000Z","dependencies_parsed_at":"2025-07-13T20:42:20.192Z","dependency_job_id":null,"html_url":"https://github.com/KempnerInstitute/nicewebrl","commit_stats":null,"previous_names":["wcarvalho/nicewebrl","kempnerinstitute/nicewebrl"],"tags_count":0,"template":false,"template_full_name":null,"purl":"pkg:github/KempnerInstitute/nicewebrl","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/KempnerInstitute%2Fnicewebrl","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/KempnerInstitute%2Fnicewebrl/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/KempnerInstitute%2Fnicewebrl/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/KempnerInstitute%2Fnicewebrl/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/KempnerInstitute","download_url":"https://codeload.github.com/KempnerInstitute/nicewebrl/tar.gz/refs/heads/main","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/KempnerInstitute%2Fnicewebrl/sbom","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":265326421,"owners_count":23747519,"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","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":"2025-06-10T02:11:05.679Z","updated_at":"2025-07-14T17:37:53.524Z","avatar_url":"https://github.com/KempnerInstitute.png","language":"Python","funding_links":[],"categories":[],"sub_categories":[],"readme":"# NiceWebRL: a Python library for human subject experiments with reinforcement learning environments\n\n\n\n**Table of Contents**\n\n* [Install](#install)\n* [Functional example folders](#working-examples)\n  * [Paper case studies](#paper-case-studies)\n  * [More examples](#more-examples)\n* [Other Jax environments compatible with NiceWebRL](#other-jax-environments-compatible-with-nicewebrl)\n\n\n\u003cimg src=\"assets/human-ai-comparisons.png\" alt=\"Comparison Image\" style=\"width: 100%; max-width: 800px;\"\u003e\n\nNiceWebRL enables researchers to use the same environments both to train and evaluate virtual agents, and to train and evaluate humans.\nIt supports both single-agent and multi-agent environments.\nAs such, NiceWebRL enables AI researchers to easily compare their algorithms to human performance, cognitive scientists to test ML algorithms as theories for human cognition, and multi-agent researchers to develop algorithms for human-AI collaboration.\n\nTo enable the use of machine learning environments in online experiments, it exploits [Jax](https://github.com/google/jax)—a high-performance numerical computing library—to reduce the latency from having clients communicate with a remote server.\nTo enable easy experiment design, NiceWebRL exploits [NiceGUI](https://nicegui.io/) to enable sophisticated GUI design entirely in Python.\n\nTo facilitate its adoption, we present several [working examples](#working-examples) that researchers can use to quickly set up their own experiments.\n## Install\n\n```bash\n# pip install\npip install git+https://github.com/wcarvalho/nicewebrl\n\n# more manually (first clone then)\nconda create -n nicewebrl python=3.10 pip wheel -y\nconda activate nicewebrl\npip install -e .\n```\n\n## Working Examples\n### Paper case studies\n\nWe present three case studies for how NiceWebRL can help researchers develop either Human-like AI, Human-compatible AI, or Human-assistive AI. The first two are from two recent papers:\n\n* [Preemptive Solving of Future Problems: Multitask Preplay in Humans and Machines](https://arxiv.org/abs/2507.05561)\n* [Cross-environment Cooperation Enables Zero-shot Multi-agent Coordination](https://arxiv.org/abs/2504.12714)\n\n\u003ctable\u003e\n  \u003cthead\u003e\n    \u003ctr\u003e\n      \u003cth\u003eDescription\u003c/th\u003e\n      \u003cth\u003eVisualization\u003c/th\u003e\n      \u003cth\u003eInstallation\u003c/th\u003e\n      \u003cth\u003eRun Command\u003c/th\u003e\n    \u003c/tr\u003e\n  \u003c/thead\u003e\n  \u003ctbody\u003e\n    \u003ctr\u003e\n      \u003ctd\u003e\u003ccode\u003eMultitask Preplay\u003c/code\u003e\u003cbr\u003e \u003cb\u003eCase study 1 (Human-like AI)\u003c/b\u003e: Developing a novel Deep RL cognitive science model with NiceWebRL [\u003ca href=\"https://arxiv.org/abs/2507.05561\"\u003ePaper\u003c/a\u003e] \u003cbr\u003e[\u003ca href=\"https://github.com/wcarvalho/multitask_preplay\"\u003efolder\u003c/a\u003e]\u003c/td\u003e\n      \u003ctd\u003e\u003cimg src=\"assets/preplay.png\" width=\"120\"/\u003e\u003c/td\u003e\n      \u003ctd\u003e\u003c/td\u003e\n      \u003ctd\u003esee \u003ca href=\"https://github.com/wcarvalho/multitask_preplay\"\u003ePaper Repo\u003c/a\u003e\u003c/td\u003e\n    \u003c/tr\u003e\n    \u003ctr\u003e\n      \u003ctd\u003e\u003ccode\u003eCross Environment Cooperation\u003c/code\u003e\u003cbr\u003e \u003cb\u003eCase study 2 (Human-compatible AI)\u003c/b\u003e: Developing a novel MARL algorithm for coordinating with humans with NiceWebRL [\u003ca href=\"https://arxiv.org/pdf/2504.12714\"\u003ePaper\u003c/a\u003e] \u003cbr\u003e[\u003ca href=\"https://github.com/wcarvalho/nicewebrl/tree/main/examples/overcooked-CEC\"\u003efolder\u003c/a\u003e]\u003c/td\u003e\n      \u003ctd\u003e\u003cimg src=\"assets/cec.png\" width=\"120\"/\u003e\u003c/td\u003e\n      \u003ctd\u003efollow instructions in \u003ccode\u003eexamples/overcooked-CEC\u003c/code\u003e\u003c/td\u003e\n      \u003ctd\u003e\u003ccode\u003ecd examples/overcooked-CEC; python web_app.py 'counter_circuit'\u003c/code\u003e\u003c/td\u003e\n    \u003c/tr\u003e\n    \u003ctr\u003e\n      \u003ctd\u003e\u003ccode\u003eXLand-Minigrid LLM\u003c/code\u003e\u003cbr\u003e \u003cb\u003eCase study 3 (Human-assistive AI)\u003c/b\u003e: Developing an LLM-assistant for sequential-decision making tasks in a virtual environment. \u003cbr\u003e[\u003ca href=\"https://github.com/wcarvalho/nicewebrl/tree/main/examples/xland-LLM-assistant\"\u003efolder\u003c/a\u003e]\u003c/td\u003e\n      \u003ctd\u003e\u003cimg src=\"assets/xland_minigrid.gif\" width=\"120\"/\u003e\u003c/td\u003e\n      \u003ctd\u003e\u003ccode\u003epip install -e \".[xland-assistant]\"\u003c/code\u003e\u003c/td\u003e\n      \u003ctd\u003e\u003ccode\u003ecd examples/xland-LLM-assistant; python web_app_assistant.py\u003c/code\u003e\u003c/td\u003e\n    \u003c/tr\u003e\n  \u003c/tbody\u003e\n\u003c/table\u003e\n\n\n### More examples\n\u003ctable\u003e\n  \u003cthead\u003e\n    \u003ctr\u003e\n      \u003cth\u003eDescription\u003c/th\u003e\n      \u003cth\u003eVisualization\u003c/th\u003e\n      \u003cth\u003eInstallation\u003c/th\u003e\n      \u003cth\u003eRun Command\u003c/th\u003e\n    \u003c/tr\u003e\n  \u003c/thead\u003e\n  \u003ctbody\u003e\n    \u003ctr\u003e\n      \u003ctd\u003e\u003ccode\u003eCraftax\u003c/code\u003e\u003cbr\u003e 2D Minecraft domain (single agent setting) \u003cbr\u003e[\u003ca href=\"https://github.com/wcarvalho/nicewebrl/tree/main/examples/craftax\"\u003efolder\u003c/a\u003e]\u003c/td\u003e\n      \u003ctd\u003e\u003cimg src=\"assets/craftax.gif\" width=\"120\"/\u003e\u003c/td\u003e\n      \u003ctd\u003e\u003ccode\u003epip install -e \".[craftax]\"\u003c/code\u003e\u003c/td\u003e\n      \u003ctd\u003e\u003ccode\u003ecd examples/craftax; python web_app.py\u003c/code\u003e\u003c/td\u003e\n    \u003c/tr\u003e\n    \u003ctr\u003e\n      \u003ctd\u003e\u003ccode\u003eXLand-Minigrid\u003c/code\u003e\u003cbr\u003e XLand-Minigrid (single agent setting) \u003cbr\u003e[\u003ca href=\"https://github.com/wcarvalho/nicewebrl/tree/main/examples/xland-minigrid\"\u003efolder\u003c/a\u003e]\u003c/td\u003e\n      \u003ctd\u003e\u003cimg src=\"assets/xland_minigrid.gif\" width=\"120\"/\u003e\u003c/td\u003e\n      \u003ctd\u003e\u003ccode\u003epip install -e \".[xland]\"\u003c/code\u003e\u003c/td\u003e\n      \u003ctd\u003e\u003ccode\u003ecd examples/xland-minigrid; python web_app.py\u003c/code\u003e\u003c/td\u003e\n    \u003c/tr\u003e\n    \u003ctr\u003e\n      \u003ctd\u003e\u003ccode\u003eMinigrid PPO\u003c/code\u003e\u003cbr\u003e Minigrid (single agent setting, has PPO implementation) \u003cbr\u003e[\u003ca href=\"https://github.com/wcarvalho/nicewebrl/tree/main/examples/minigrid-ppo\"\u003efolder\u003c/a\u003e]\u003c/td\u003e\n      \u003ctd\u003e\u003cimg src=\"assets/navix.gif\" width=\"120\"/\u003e\u003c/td\u003e\n      \u003ctd\u003e\u003ccode\u003epip install -e \".\"\u003c/code\u003e\u003c/td\u003e\n      \u003ctd\u003e\u003ccode\u003ecd examples/minigrid-ppo; python web_app.py\u003c/code\u003e\u003c/td\u003e\n    \u003c/tr\u003e\n    \u003ctr\u003e\n      \u003ctd\u003e\u003ccode\u003eMinigrid API LLM\u003c/code\u003e\u003cbr\u003e Minigrid with API-based LLM assistant \u003cbr\u003e[\u003ca href=\"https://github.com/wcarvalho/nicewebrl/tree/main/examples/minigrid-LLM-assistant\"\u003efolder\u003c/a\u003e]\u003c/td\u003e\n      \u003ctd\u003e\u003cimg src=\"assets/navix.gif\" width=\"120\"/\u003e\u003c/td\u003e\n      \u003ctd\u003e\u003ccode\u003epip install -e \".\"\u003c/code\u003e\u003c/td\u003e\n      \u003ctd\u003e\u003ccode\u003ecd examples/minigrid-LLM-assistant; python web_app_Gemini.py\u003c/code\u003e\u003c/td\u003e\n    \u003c/tr\u003e\n    \u003ctr\u003e\n      \u003ctd\u003e\u003ccode\u003eMinigrid Local LLM\u003c/code\u003e\u003cbr\u003e Minigrid with \u003cb\u003elocal\u003c/b\u003e LLM assistant \u003cbr\u003e[\u003ca href=\"https://github.com/wcarvalho/nicewebrl/tree/main/examples/minigrid-LLM-assistant\"\u003efolder\u003c/a\u003e]\u003c/td\u003e\n      \u003ctd\u003e\u003cimg src=\"assets/navix.gif\" width=\"120\"/\u003e\u003c/td\u003e\n      \u003ctd\u003e\u003ccode\u003epip install -e \".\"\u003c/code\u003e\u003c/td\u003e\n      \u003ctd\u003e\u003ccode\u003ecd examples/minigrid-LLM-assistant; python web_app_Local.py\u003c/code\u003e\u003c/td\u003e\n    \u003c/tr\u003e\n    \u003ctr\u003e\n      \u003ctd\u003e\u003ccode\u003eDual Destination\u003c/code\u003e\u003cbr\u003e Dual Destination (Human-AI setting) \u003cbr\u003e[\u003ca href=\"https://github.com/wcarvalho/nicewebrl/tree/main/examples/dual_destination-human-AI-CEC\"\u003efolder\u003c/a\u003e]\u003c/td\u003e\n      \u003ctd\u003e\u003cimg src=\"assets/dual-destination.png\" width=\"120\"/\u003e\u003c/td\u003e\n      \u003ctd\u003e\u003ccode\u003epip install -e \".[jaxmarl]\"\u003c/code\u003e\u003c/td\u003e\n      \u003ctd\u003e\u003ccode\u003ecd examples/dual_destination-human-AI-CEC; python web_app.py\u003c/code\u003e\u003c/td\u003e\n    \u003c/tr\u003e\n    \u003ctr\u003e\n      \u003ctd\u003e\u003ccode\u003eOvercooked\u003c/code\u003e\u003cbr\u003e Overcooked (Human-Human setting) \u003cbr\u003e[\u003ca href=\"https://github.com/wcarvalho/nicewebrl/tree/main/examples/overcooked-human-human\"\u003efolder\u003c/a\u003e]\u003c/td\u003e\n      \u003ctd\u003e\u003cimg src=\"assets/jaxmarl.gif\" width=\"120\"/\u003e\u003c/td\u003e\n      \u003ctd\u003e\u003ccode\u003epip install -e \".[jaxmarl]\"\u003c/code\u003e\u003c/td\u003e\n      \u003ctd\u003e\u003ccode\u003ecd examples/overcooked-human-human; python web_app.py\u003c/code\u003e\u003c/td\u003e\n    \u003c/tr\u003e\n  \u003c/tbody\u003e\n\u003c/table\u003e\n\n\n## Other Jax environments compatible with NiceWebRL\n\nThe following are all Jax environments which can be used with this framework to run human subject experiments. The power of using jax is that one can use the **exact** same environment for human subjects experiments as for developing modern machine learning algorithms (especially reinforcement learning algorithms).\n\nWhen targetting normative solutions, one may want to study algorithms asymptotic behavior with a lot of data. Jax makes it cheap to do this. NiceWebRL makes it easy to compare these algorithms to human subject behavior.\n\u003c!--\u003ctable style=\"width:100%; border-collapse: collapse;\"\u003e\n  \u003ctr style=\"max-height: 150px; overflow: hidden;\"\u003e\n    \u003ctd style=\"border: 1px solid black; padding: 10px; text-align: center;\"\u003e\n      \u003ca href=\"https://github.com/MichaelTMatthews/Craftax\" target=\"_blank\" style=\"text-decoration: none; color: inherit;\"\u003e\n        \u003ccenter\u003e\u003cstrong\u003eCraftax\u003c/strong\u003e\u003c/center\u003e\n      \u003c/a\u003e\u003cbr\u003e\n      \u003ca href=\"https://github.com/MichaelTMatthews/Craftax\" target=\"_blank\"\u003e\n        \u003cimg src=\"https://raw.githubusercontent.com/MichaelTMatthews/Craftax/main/images/building.gif\" alt=\"Craftax\" style=\"width: 100%; max-width: 400px;\"\u003e\n      \u003c/a\u003e\n      \u003cp\u003eThis is a grid-world version of minecraft. \u003c/p\u003e\n    \u003c/td\u003e\n    \u003ctd style=\"border: 1px solid black; padding: 10px; text-align: center;\"\u003e\n      \u003ca href=\"https://github.com/wcarvalho/JaxHouseMaze\" target=\"_blank\" style=\"text-decoration: none; color: inherit;\"\u003e\n        \u003ccenter\u003e\u003cstrong\u003eHousemaze\u003c/strong\u003e\u003c/center\u003e\n      \u003c/a\u003e\u003cbr\u003e\n      \u003ca href=\"https://github.com/wcarvalho/JaxHouseMaze\" target=\"_blank\"\u003e\n        \u003cimg src=\"https://github.com/wcarvalho/JaxHouseMaze/raw/main/example.png\" alt=\"Housemaze\" style=\"width: 100%; max-width: 400px;\"\u003e\n      \u003c/a\u003e\n      \u003cp\u003eThis is a maze environment where new mazes can be easily be described with a string.\u003c/p\u003e\n    \u003c/td\u003e\n    \u003ctd style=\"border: 1px solid black; padding: 10px; text-align: center;\"\u003e\n      \u003ca href=\"https://github.com/corl-team/xland-minigrid\" target=\"_blank\" style=\"text-decoration: none; color: inherit;\"\u003e\n        \u003ccenter\u003e\u003cstrong\u003eXLand-Minigrid\u003c/strong\u003e\u003c/center\u003e\n      \u003c/a\u003e\u003cbr\u003e\n      \u003ca href=\"https://github.com/corl-team/xland-minigrid\" target=\"_blank\"\u003e\n        \u003cimg src=\"https://github.com/corl-team/xland-minigrid/blob/main/figures/ruleset-example.jpg?raw=true\" alt=\"XLand-Minigrid\" style=\"width: 100%; max-width: 400px;\"\u003e\n      \u003c/a\u003e\n      \u003cp\u003eThis environment allows for complex, nested compositional tasks. XLand-Minigrid comes with 3 benchmarks which together defnine 3 million unique tasks.\u003c/p\u003e\n    \u003c/td\u003e\n  \u003c/tr\u003e\n  \u003ctr style=\"max-height: 150px; overflow: hidden;\"\u003e\n    \u003ctd style=\"border: 1px solid black; padding: 10px; text-align: center;\"\u003e\n      \u003ca href=\"https://github.com/epignatelli/navix\" target=\"_blank\" style=\"text-decoration: none; color: inherit;\"\u003e\n        \u003ccenter\u003e\u003cstrong\u003eNavix\u003c/strong\u003e\u003c/center\u003e\n      \u003c/a\u003e\u003cbr\u003e\n      \u003ca href=\"https://github.com/epignatelli/navix\" target=\"_blank\"\u003e\n        \u003cimg src=\"https://minigrid.farama.org/_images/GoToObjectEnv.gif\" alt=\"Navix\" style=\"width: 100%; max-width: 400px;\"\u003e\n      \u003c/a\u003e\n      \u003cp\u003eThis is a jax implementation of the popular Minigrid environment.\u003c/p\u003e\n    \u003c/td\u003e\n    \u003ctd style=\"border: 1px solid black; padding: 10px; text-align: center;\"\u003e\n      \u003ca href=\"https://github.com/FLAIROx/JaxMARL\" target=\"_blank\" style=\"text-decoration: none; color: inherit;\"\u003e\n        \u003ccenter\u003e\u003cstrong\u003eOvercooked (multi-agent)\u003c/strong\u003e\u003c/center\u003e\n      \u003c/a\u003e\u003cbr\u003e\n      \u003ca href=\"https://github.com/FLAIROx/JaxMARL\" target=\"_blank\"\u003e\n        \u003cimg src=\"https://github.com/FLAIROx/JaxMARL/blob/main/docs/imgs/cramped_room.gif?raw=true\" alt=\"Overcooked\" style=\"width: 100%; max-width: 400px;\"\u003e\n      \u003c/a\u003e\n      \u003cp\u003eThis is a popular multi-agent environment.\u003c/p\u003e\n    \u003c/td\u003e\n    \u003ctd style=\"border: 1px solid black; padding: 10px; text-align: center;\"\u003e\n      \u003ca href=\"https://github.com/FLAIROx/JaxMARL\" target=\"_blank\" style=\"text-decoration: none; color: inherit;\"\u003e\n        \u003ccenter\u003e\u003cstrong\u003eSTORM (multi-agent)\u003c/strong\u003e\u003c/center\u003e\n      \u003c/a\u003e\u003cbr\u003e\n      \u003ca href=\"https://github.com/FLAIROx/JaxMARL\" target=\"_blank\"\u003e\n        \u003cimg src=\"https://github.com/FLAIROx/JaxMARL/blob/main/docs/imgs/storm.gif?raw=true\" alt=\"STORM\" style=\"width: 100%; max-width: 400px;\"\u003e\n      \u003c/a\u003e\n      \u003cp\u003eThis environment specifies Matrix games represented as grid world scenarios.\u003c/p\u003e\n    \u003c/td\u003e\n  \u003c/tr\u003e\n\u003c/table\u003e\n--\u003e\n\n\n\n\u003ctable\u003e\n  \u003cthead\u003e\n    \u003ctr\u003e\n      \u003cth\u003eName\u003c/th\u003e\n      \u003cth\u003eDomain\u003c/th\u003e\n      \u003cth\u003eVisualization\u003c/th\u003e\n      \u003cth\u003eGoal\u003c/th\u003e\n    \u003c/tr\u003e\n  \u003c/thead\u003e\n  \u003ctbody\u003e\n    \u003ctr\u003e\n      \u003ctd\u003e\u003ccode\u003eCraftax\u003c/code\u003e\u003c/td\u003e\n      \u003ctd\u003e2D Minecraft\u003c/td\u003e\n      \u003ctd\u003e\u003cimg src=\"assets/craftax.gif\" width=\"120\"/\u003e\u003c/td\u003e\n      \u003ctd\u003eMine and craft resources in a Minecraft-like 2D world\u003c/td\u003e\n    \u003c/tr\u003e\n    \u003ctr\u003e\n      \u003ctd\u003e\u003ccode\u003eKinetix\u003c/code\u003e\u003c/td\u003e\n      \u003ctd\u003e2D Physics Control\u003c/td\u003e\n      \u003ctd\u003e\u003cimg src=\"assets/kinetix.gif\" width=\"120\"/\u003e\u003c/td\u003e\n      \u003ctd\u003eControl rigid 2D bodies to perform dynamic tasks\u003c/td\u003e\n    \u003c/tr\u003e\n    \u003ctr\u003e\n      \u003ctd\u003e\u003ccode\u003eNavix\u003c/code\u003e\u003c/td\u003e\n      \u003ctd\u003eMiniGrid\u003c/td\u003e\n      \u003ctd\u003e\u003cimg src=\"assets/navix.gif\" width=\"120\"/\u003e\u003c/td\u003e\n      \u003ctd\u003eNavigate grid environments with JAX-based MiniGrid variant\u003c/td\u003e\n    \u003c/tr\u003e\n    \u003ctr\u003e\n      \u003ctd\u003e\u003ccode\u003eXLand–MiniGrid\u003c/code\u003e\u003c/td\u003e\n      \u003ctd\u003eXLand\u003c/td\u003e\n      \u003ctd\u003e\u003cimg src=\"assets/xland_minigrid.gif\" width=\"120\"/\u003e\u003c/td\u003e\n      \u003ctd\u003eMeta-RL tasks defined by dynamic rulesets\u003c/td\u003e\n    \u003c/tr\u003e\n    \u003ctr\u003e\n      \u003ctd\u003e\u003ccode\u003eJaxMARL\u003c/code\u003e\u003c/td\u003e\n      \u003ctd\u003eMulti-agent RL\u003c/td\u003e\n      \u003ctd\u003e\u003cimg src=\"assets/jaxmarl.gif\" width=\"120\"/\u003e\u003c/td\u003e\n      \u003ctd\u003eCooperative and competitive multi-agent environments in JAX\u003c/td\u003e\n    \u003c/tr\u003e\n    \u003ctr\u003e\n      \u003ctd\u003e\u003ccode\u003eJaxGCRL\u003c/code\u003e\u003c/td\u003e\n      \u003ctd\u003eGoal-Conditioned Robotics\u003c/td\u003e\n      \u003ctd\u003e\u003cimg src=\"assets/jaxgcrl.gif\" width=\"120\"/\u003e\u003c/td\u003e\n      \u003ctd\u003eGoal-conditioned control in simulated robotics tasks\u003c/td\u003e\n    \u003c/tr\u003e\n    \u003ctr\u003e\n      \u003ctd\u003e\u003ccode\u003eGymnax\u003c/code\u003e\u003c/td\u003e\n      \u003ctd\u003eClassic RL\u003c/td\u003e\n      \u003ctd\u003e\u003cimg src=\"assets/gymnax.gif\" width=\"120\"/\u003e\u003c/td\u003e\n      \u003ctd\u003eClassic control, bsuite, and MinAtar environments in JAX\u003c/td\u003e\n    \u003c/tr\u003e\n    \u003ctr\u003e\n      \u003ctd\u003e\u003ccode\u003eJumanji\u003c/code\u003e\u003c/td\u003e\n      \u003ctd\u003eCombinatorial\u003c/td\u003e\n      \u003ctd\u003e\u003cimg src=\"assets/jumanji.gif\" width=\"120\"/\u003e\u003c/td\u003e\n      \u003ctd\u003eFrom simple games to NP-hard combinatorial problems\u003c/td\u003e\n    \u003c/tr\u003e\n    \u003ctr\u003e\n      \u003ctd\u003e\u003ccode\u003ePgx\u003c/code\u003e\u003c/td\u003e\n      \u003ctd\u003eBoard Games\u003c/td\u003e\n      \u003ctd\u003e\u003cimg src=\"assets/pgx.gif\" width=\"120\"/\u003e\u003c/td\u003e\n      \u003ctd\u003eChess, Go, Shogi, and other turn-based strategy games\u003c/td\u003e\n    \u003c/tr\u003e\n    \u003ctr\u003e\n      \u003ctd\u003e\u003ccode\u003eBrax\u003c/code\u003e\u003c/td\u003e\n      \u003ctd\u003ePhysics Simulation\u003c/td\u003e\n      \u003ctd\u003e\u003cimg src=\"assets/brax.gif\" width=\"120\"/\u003e\u003c/td\u003e\n      \u003ctd\u003eDifferentiable physics engine for continuous control\u003c/td\u003e\n    \u003c/tr\u003e\n  \u003c/tbody\u003e\n\u003c/table\u003e\n\n[![Star History Chart](https://api.star-history.com/svg?repos=KempnerInstitute/nicewebrl\u0026type=Date)](https://star-history.com/#KempnerInstitute/nicewebrl\u0026Date)\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fkempnerinstitute%2Fnicewebrl","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fkempnerinstitute%2Fnicewebrl","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fkempnerinstitute%2Fnicewebrl/lists"}