{"id":24483148,"url":"https://github.com/sukiboo/sxf","last_synced_at":"2026-05-14T05:33:08.485Z","repository":{"id":223190134,"uuid":"759533923","full_name":"sukiboo/sxf","owner":"sukiboo","description":"Synthetic Experimental Framework for hyper-personalized behavioral nudging","archived":false,"fork":false,"pushed_at":"2024-02-22T15:14:41.000Z","size":20,"stargazers_count":0,"open_issues_count":0,"forks_count":0,"subscribers_count":2,"default_branch":"main","last_synced_at":"2025-01-21T12:35:52.539Z","etag":null,"topics":["experimental-design","nudging","reinforcement-learning","simulated-environments","tensorflow"],"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/sukiboo.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}},"created_at":"2024-02-18T21:02:31.000Z","updated_at":"2024-02-18T21:08:48.000Z","dependencies_parsed_at":"2024-02-18T22:23:31.711Z","dependency_job_id":"cefb8008-9349-45e9-af5e-ecd6328b48be","html_url":"https://github.com/sukiboo/sxf","commit_stats":null,"previous_names":["sukiboo/sxf"],"tags_count":0,"template":false,"template_full_name":null,"repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/sukiboo%2Fsxf","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/sukiboo%2Fsxf/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/sukiboo%2Fsxf/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/sukiboo%2Fsxf/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/sukiboo","download_url":"https://codeload.github.com/sukiboo/sxf/tar.gz/refs/heads/main","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":243639404,"owners_count":20323505,"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":["experimental-design","nudging","reinforcement-learning","simulated-environments","tensorflow"],"created_at":"2025-01-21T12:31:04.819Z","updated_at":"2026-05-14T05:33:03.431Z","avatar_url":"https://github.com/sukiboo.png","language":"Python","funding_links":[],"categories":[],"sub_categories":[],"readme":"# Synthetic Experimental Framework\n\nThis repository includes a customizable simulated environment for testing different aspects of the agent/environment interaction in addressing the behavioral nudging personalization contextual bandit problem. The code was written around 2022 and may be a bit outdated now.\n\nThe findings from SXF contributed to the papers\n- [Examining Policy Entropy of Reinforcement Learning Agents for Personalization Tasks](https://arxiv.org/abs/2211.11869)\n- [Increasing Entropy to Boost Policy Gradient Performance on Personalization Tasks](https://arxiv.org/abs/2310.05324)\n- [On the Unreasonable Efficiency of State Space Clustering in Personalization Tasks](https://arxiv.org/abs/2112.13141)\n\n![exp_action_dist](https://github.com/sukiboo/sxf/assets/38059493/931978ba-2d8c-43f4-b3de-01704f24f358)\n\n## Installation\n* Install conda / pip requirements via `conda env create -f environment.yml`\n* Activate conda environment with `conda activate synthetic-experimental-framework`\n* Modify experiment configuration in `configs/config.yml` as needed\n* Run a new experiment via `python -m run_experiment`, or load a recorded experiment `exp_name` via `python -m run_experiment --load exp_name`\n* Results of the experiment can be found in `exp_data/exp_name/` directory\n\n## File Overview\n* `environment.yml` --- list of the required packages\n* `configs/config.yml` --- config file containing the experiment/envirnment/agent parameters\n* `run_experiment.py` --- main module to run the experiment\n* `exp_data/` --- directory containing the experiment data, images, checkpoints\n---\n* `experiment_component/experiment.py` --- setup the experiment\n* `experiment_component/data_visualization.py` --- compute and report results of an experiment\n---\n* `environment_component/environment.py` --- setup the environment\n* `environment_component/state_space.py` --- setup the state space for the environment\n* `environment_component/action_space.py` --- setup the action space for the environment\n* `environment_component/reward_function.py` --- setup the reward function for the environment\n* `environment_component/feedback_signal.py` --- setup the feedback signal that is given to the agent\n---\n* `agent_component/agent.py` --- setup the agent\n* `agent_component/network_architecture.py` --- setup the policy for the agent\n* `agent_component/loss_function.py` --- setup the loss function for the agent\n\n\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fsukiboo%2Fsxf","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fsukiboo%2Fsxf","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fsukiboo%2Fsxf/lists"}