{"id":19401027,"url":"https://github.com/google-research/recsim_ng","last_synced_at":"2025-04-24T07:30:27.313Z","repository":{"id":49373254,"uuid":"329109817","full_name":"google-research/recsim_ng","owner":"google-research","description":"RecSim NG: Toward Principled Uncertainty Modeling for Recommender Ecosystems","archived":false,"fork":false,"pushed_at":"2022-04-26T19:56:53.000Z","size":3874,"stargazers_count":118,"open_issues_count":5,"forks_count":16,"subscribers_count":7,"default_branch":"master","last_synced_at":"2025-04-18T18:33:28.538Z","etag":null,"topics":["artificial-intelligence","google","probabilistic-programming","recommender-system","reinforcement-learning","simulation","tensorflow"],"latest_commit_sha":null,"homepage":"https://github.com/google-research/recsim_ng","language":"Jupyter Notebook","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"apache-2.0","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/google-research.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":"CONTRIBUTING.md","funding":null,"license":"LICENSE","code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null}},"created_at":"2021-01-12T20:54:15.000Z","updated_at":"2025-01-31T05:48:15.000Z","dependencies_parsed_at":"2022-09-01T01:12:51.063Z","dependency_job_id":null,"html_url":"https://github.com/google-research/recsim_ng","commit_stats":null,"previous_names":[],"tags_count":0,"template":false,"template_full_name":null,"repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/google-research%2Frecsim_ng","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/google-research%2Frecsim_ng/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/google-research%2Frecsim_ng/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/google-research%2Frecsim_ng/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/google-research","download_url":"https://codeload.github.com/google-research/recsim_ng/tar.gz/refs/heads/master","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":250582772,"owners_count":21453911,"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":["artificial-intelligence","google","probabilistic-programming","recommender-system","reinforcement-learning","simulation","tensorflow"],"created_at":"2024-11-10T11:16:46.641Z","updated_at":"2025-04-24T07:30:26.657Z","avatar_url":"https://github.com/google-research.png","language":"Jupyter Notebook","readme":"# RecSim NG: Toward Principled Uncertainty Modeling for Recommender Ecosystems\n\nRecSim NG, a probabilistic platform for multi-agent recommender systems\nsimulation. RecSimNG is a scalable, modular, differentiable simulator\nimplemented in Edward2 and TensorFlow. It offers: a powerful, general\nprobabilistic programming language for agent-behavior specification; an\nXLA-based vectorized execution model for running simulations on accelerated\nhardware; and tools for probabilistic inference and latent-variable model\nlearning, backed by automatic differentiation and tracing. We describe RecSim NG\nand illustrate how it can be used to create transparent, configurable,\nend-to-end models of a recommender ecosystem. Specifically, we present a\ncollection of use cases that demonstrate how the functionality described above\ncan help both researchers and practitioners easily develop and train novel\nalgorithms for recommender systems. Please refer to\n[Mladenov et al](https://arxiv.org/abs/2103.08057) for the\nhigh-level design of RecSim NG. Please cite the paper if you use the code from\nthis repository in your work.\n\n### Bibtex\n\n```\n@article{mladenov2021recsimng,\n    title = {RecSim {NG}: Toward Principled Uncertainty Modeling for Recommender Ecosystems},\n    author = {Martin Mladenov, Chih-Wei Hsu, Vihan Jain, Eugene Ie, Christopher Colby, Nicolas Mayoraz, Hubert Pham, Dustin Tran, Ivan Vendrov, Craig Boutilier}\n    year = {2021},\n    eprint={2103.08057},\n    archivePrefix={arXiv},\n    primaryClass={cs.LG}\n}\n```\n\n\u003ca id='Disclaimer'\u003e\u003c/a\u003e\n\n## Disclaimer\n\nThis is not an officially supported Google product.\n\n## Installation and Sample Usage\n\nIt is recommended to install RecSim NG using\n(https://pypi.org/project/recsim_ng).\n\n```shell\npip install recsim_ng\n```\n\nHere are some sample commands you could use for testing the installation:\n\n```\ngit clone https://github.com/google-research/recsim_ng\ncd recsim_ng/recsim_ng/applications/ecosystem_simulation\npython ecosystem_simulation_demo.py\n```\n\n## Tutorials\n\nTo get started, please check out our Colab tutorials. In\n[**RecSim NG: Basics**](https://colab.research.google.com/github/google-research/recsim_ng/blob/master/recsim_ng/colab/RecSim_NG_Basics.ipynb),\nwe introduce the RecSim NG model and corresponding modeling APIs and runtime\nlibrary. We then demonstrate how we define a simulation using **entities**,\n**behaviors**, and **stories**. Finally, we illustrate differentiable\nsimulation including model learning and inference.\n\nIn [**RecSim NG: Dealing With Uncertainty**](https://colab.research.google.com/github/google-research/recsim_ng/blob/master/recsim_ng/colab/RecSim_NG_Dealing_With_Uncertainty.ipynb),\nwe explicitly address the stochastics of the Markov process captured by a DBN.\nWe demonstrate how to use Edward2 in RecSim NG and show how to use the\ncorresponding RecSim NG APIs for inference and learning tasks. Finally, we\nshowcase how the uncertainty APIs of RecSim NG can be used within a\nrecommender-system model-learning application.\n\n## Documentation\n\n\nPlease refer to the [white paper](https://arxiv.org/abs/2103.08057)\nfor the high-level design.\n","funding_links":[],"categories":[],"sub_categories":[],"project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fgoogle-research%2Frecsim_ng","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fgoogle-research%2Frecsim_ng","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fgoogle-research%2Frecsim_ng/lists"}