{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/phasic-policy-gradient","title":"Phasic Policy Gradient","arxiv_id":"2009.04416","date":"2020-09-09","proceeding":null,"authors":["Karl Cobbe","Jacob Hilton","Oleg Klimov","John Schulman"],"abstract":"We introduce Phasic Policy Gradient (PPG), a reinforcement learning framework which modifies traditional on-policy actor-critic methods by separating policy and value function training into distinct phases. In prior methods, one must choose between using a shared network or separate networks to represent the policy and value function. Using separate networks avoids interference between objectives, while using a shared network allows useful features to be shared. PPG is able to achieve the best of both worlds by splitting optimization into two phases, one that advances training and one that distills features. PPG also enables the value function to be more aggressively optimized with a higher level of sample reuse. Compared to PPO, we find that PPG significantly improves sample efficiency on the challenging Procgen Benchmark.","url_abs":"https://arxiv.org/abs/2009.04416v1","url_pdf":"https://arxiv.org/pdf/2009.04416v1.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"phasic-policy-gradient","repo_url":"https://github.com/openai/phasic-policy-gradient","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"MIT"}},{"paper_slug":"phasic-policy-gradient","repo_url":"https://github.com/jjccero/pbrl/tree/master/pbrl/algorithms/ppg","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"pytorch","reach":{"status":"ok","spdx":"MIT"}},{"paper_slug":"phasic-policy-gradient","repo_url":"https://github.com/opendilab/DI-engine/blob/main/ding/policy/ppg.py","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"reinforcement-learning-1","task_name":"Reinforcement Learning (RL)"}],"methods":[{"method_slug":"entropy-regularization","method_name":"Entropy Regularization"},{"method_slug":"ppo","method_name":"PPO"}],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/reinforcement-learning-on-procgen","task":"Reinforcement Learning (RL)","dataset":"ProcGen","model":"PPG","rank_in_archive_order":1,"of":2,"metrics":{"Mean Normalized Performance":"0.757"},"uses_additional_data":true},{"leaderboard":"/sota/reinforcement-learning-on-procgen","task":"Reinforcement Learning (RL)","dataset":"ProcGen","model":"PPO","rank_in_archive_order":2,"of":2,"metrics":{"Mean Normalized Performance":"0.576"},"uses_additional_data":true}],"syntology":{"syntology_url":"https://syntology.ai/paper/2009.04416","atlas_url":"https://app.syntology.ai/?focus=2009.04416","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2009.04416"}},"developers":"https://syntology.ai/developers","read_at":"2026-09-25T09:33:49+00:00","read_at_is":"when the build read Syntology's graph, not when any sample ran","claim":"Per-sample execution status on synthesized fixtures; not a correctness claim about the paper. Samples come from repositories linked to the paper, official or community; repo_kind says which.","repos":[{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/opendilab/DI-engine/blob/main/ding/policy/ppg.py","reach":null},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/jjccero/pbrl/tree/master/pbrl/algorithms/ppg","reach":{"status":"ok","spdx":"MIT"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/openai/phasic-policy-gradient","reach":{"status":"ok","spdx":"MIT"}}],"summary":{"unverified":13},"by_repo_kind":{"official":{"samples":13,"ran":0,"repositories":1}},"repo_kind_vocabulary":{"official":"The archive marks this repository official for the paper","named_in_paper":"The archive records that the paper mentions this repository; it is not marked official","listed":"In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper","found_in_text":"Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted","community":"Not in the archive's code links for this paper; a community repository Syntology harvested"},"n_pointer_only_for_licence":0,"samples":[{"code_sha256_prefix":"6d9e098f75a09f80","entry":"compute_losses","repo":"openai/phasic-policy-gradient","repo_kind":"official","path":"phasic_policy_gradient/ppo.py","file_url":"https://github.com/openai/phasic-policy-gradient/blob/HEAD/phasic_policy_gradient/ppo.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"6d9e098f75a09f80"}},{"code_sha256_prefix":"80025f416c031ad6","entry":"dict_mean","repo":"openai/phasic-policy-gradient","repo_kind":"official","path":"phasic_policy_gradient/minibatch_optimize.py","file_url":"https://github.com/openai/phasic-policy-gradient/blob/HEAD/phasic_policy_gradient/minibatch_optimize.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"80025f416c031ad6"}},{"code_sha256_prefix":"a362c435a091c0f4","entry":"ema","repo":"openai/phasic-policy-gradient","repo_kind":"official","path":"phasic_policy_gradient/graph_util.py","file_url":"https://github.com/openai/phasic-policy-gradient/blob/HEAD/phasic_policy_gradient/graph_util.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"a362c435a091c0f4"}},{"code_sha256_prefix":"6b6c572969663aab","entry":"make_output_format","repo":"openai/phasic-policy-gradient","repo_kind":"official","path":"phasic_policy_gradient/logger.py","file_url":"https://github.com/openai/phasic-policy-gradient/blob/HEAD/phasic_policy_gradient/logger.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"6b6c572969663aab"}},{"code_sha256_prefix":"98d06afb4dcd2228","entry":"minibatch_optimize","repo":"openai/phasic-policy-gradient","repo_kind":"official","path":"phasic_policy_gradient/minibatch_optimize.py","file_url":"https://github.com/openai/phasic-policy-gradient/blob/HEAD/phasic_policy_gradient/minibatch_optimize.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"98d06afb4dcd2228"}},{"code_sha256_prefix":"fc70defdf687afcb","entry":"mpi_weighted_mean","repo":"openai/phasic-policy-gradient","repo_kind":"official","path":"phasic_policy_gradient/logger.py","file_url":"https://github.com/openai/phasic-policy-gradient/blob/HEAD/phasic_policy_gradient/logger.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"fc70defdf687afcb"}},{"code_sha256_prefix":"ca2c1a746094ca28","entry":"profile","repo":"openai/phasic-policy-gradient","repo_kind":"official","path":"phasic_policy_gradient/logger.py","file_url":"https://github.com/openai/phasic-policy-gradient/blob/HEAD/phasic_policy_gradient/logger.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"ca2c1a746094ca28"}},{"code_sha256_prefix":"fe74ca9ccc423b88","entry":"rcm","repo":"openai/phasic-policy-gradient","repo_kind":"official","path":"phasic_policy_gradient/log_save_helper.py","file_url":"https://github.com/openai/phasic-policy-gradient/blob/HEAD/phasic_policy_gradient/log_save_helper.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"fe74ca9ccc423b88"}},{"code_sha256_prefix":"ed1d9f7bf3690847","entry":"read_csv","repo":"openai/phasic-policy-gradient","repo_kind":"official","path":"phasic_policy_gradient/graph_util.py","file_url":"https://github.com/openai/phasic-policy-gradient/blob/HEAD/phasic_policy_gradient/graph_util.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"ed1d9f7bf3690847"}},{"code_sha256_prefix":"8b47a7ede19e9b17","entry":"sum_nonbatch","repo":"openai/phasic-policy-gradient","repo_kind":"official","path":"phasic_policy_gradient/ppg.py","file_url":"https://github.com/openai/phasic-policy-gradient/blob/HEAD/phasic_policy_gradient/ppg.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"8b47a7ede19e9b17"}},{"code_sha256_prefix":"a080ad6d9c26a9e9","entry":"switch_to_outer_plot","repo":"openai/phasic-policy-gradient","repo_kind":"official","path":"phasic_policy_gradient/graph_util.py","file_url":"https://github.com/openai/phasic-policy-gradient/blob/HEAD/phasic_policy_gradient/graph_util.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"a080ad6d9c26a9e9"}},{"code_sha256_prefix":"f73c4143235d17d6","entry":"to_th_device","repo":"openai/phasic-policy-gradient","repo_kind":"official","path":"phasic_policy_gradient/minibatch_optimize.py","file_url":"https://github.com/openai/phasic-policy-gradient/blob/HEAD/phasic_policy_gradient/minibatch_optimize.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"f73c4143235d17d6"}},{"code_sha256_prefix":"d501c1a9124aff24","entry":"update_mean_var_count_from_moments","repo":"openai/phasic-policy-gradient","repo_kind":"official","path":"phasic_policy_gradient/reward_normalizer.py","file_url":"https://github.com/openai/phasic-policy-gradient/blob/HEAD/phasic_policy_gradient/reward_normalizer.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"d501c1a9124aff24"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}