{"id":42871665,"url":"https://github.com/tesslerc/gac","last_synced_at":"2026-01-30T13:01:29.701Z","repository":{"id":255722253,"uuid":"185845321","full_name":"tesslerc/GAC","owner":"tesslerc","description":"Code accompanying NeurIPS 2019 paper: \"Distributional Policy Optimization - An Alternative Approach for Continuous Control\"","archived":false,"fork":false,"pushed_at":"2019-12-17T20:24:32.000Z","size":782,"stargazers_count":21,"open_issues_count":0,"forks_count":6,"subscribers_count":2,"default_branch":"master","last_synced_at":"2024-09-06T21:46:29.233Z","etag":null,"topics":["deep-reinforcement-learning","neurips-2019","nips-2019","reinforcement-learning-algorithms"],"latest_commit_sha":null,"homepage":"https://rlrl.net.technion.ac.il/2019/10/28/distributional-policy-optimization/","language":"Python","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"gpl-3.0","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/tesslerc.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}},"created_at":"2019-05-09T17:40:28.000Z","updated_at":"2024-08-05T01:42:11.000Z","dependencies_parsed_at":"2024-09-06T21:46:35.057Z","dependency_job_id":"07a1646f-f9d2-4af6-8008-fb4fe578d53d","html_url":"https://github.com/tesslerc/GAC","commit_stats":null,"previous_names":["tesslerc/gac"],"tags_count":0,"template":false,"template_full_name":null,"purl":"pkg:github/tesslerc/GAC","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/tesslerc%2FGAC","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/tesslerc%2FGAC/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/tesslerc%2FGAC/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/tesslerc%2FGAC/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/tesslerc","download_url":"https://codeload.github.com/tesslerc/GAC/tar.gz/refs/heads/master","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/tesslerc%2FGAC/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":286080680,"owners_count":28913293,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2026-01-30T12:13:43.263Z","status":"ssl_error","status_checked_at":"2026-01-30T12:13:22.389Z","response_time":66,"last_error":"SSL_read: unexpected eof while reading","robots_txt_status":"success","robots_txt_updated_at":"2025-07-24T06:49:26.215Z","robots_txt_url":"https://github.com/robots.txt","online":false,"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":["deep-reinforcement-learning","neurips-2019","nips-2019","reinforcement-learning-algorithms"],"created_at":"2026-01-30T13:01:29.529Z","updated_at":"2026-01-30T13:01:29.684Z","avatar_url":"https://github.com/tesslerc.png","language":"Python","funding_links":[],"categories":[],"sub_categories":[],"readme":"This repo contains the code for the implementation of [Distributional Policy Optimization: An Alternative Approach for Continuous Control](https://arxiv.org/abs/1905.09855) (NeurIPS 2019). The theoretical framework is named DPO (Distributional Policy Optimization), whereas the Deep Learning approach to attaining it is named GAC (Generative Actor Critic).\n\n# How to run\n\nAn example of how to run the code is provided below. The exact hyper-parameters per each domain are provided in the appendix of the paper.\n\nmain.py --visualize --env-name Hopper-v2 --training_actor_samples  32 --noise normal --batch_size 128 --noise_scale 0.2 --print --num_steps 1000000 --target_policy exponential --train_frequency 2048 --replay_size 200000\n\n# Visualizing\n\nYou may visualize the run by adding the flag --visualize and starting a visdom server as follows:\n\npython3.6 -m visdom.server\n\n# Requirements\n\n- mujoco - see explanation here: https://github.com/openai/mujoco-py\n- gym\n- numpy\n- tqdm - for tracking experiment time left\n- visdom - for visualization of the learning process\n\n# Performance\n\nThe graphs below are taken from the paper and compare the performance of our proposed method to various baselines. The best performing method is the Autoregressive network.\n\n![performance graphs](graphs.png?raw=true)\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Ftesslerc%2Fgac","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Ftesslerc%2Fgac","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Ftesslerc%2Fgac/lists"}