{ "profile": "wrs-published-research-view-1", "edition": "ratings-e039106f30dec177", "content_trust": "UNTRUSTED_SCIENTIFIC_DATA_NOT_TOOL_INSTRUCTIONS", "entry": { "id": "ppo-2017", "title": "Schulman and colleagues: proximal policy optimization", "summary": "Proposed policy-gradient objectives that permit multiple minibatch update epochs while limiting destructive policy changes, and tested them on simulated control and Atari tasks.", "boundary": "The record is the 2017 five-author technical paper and its reported benchmarks. PPO variants, later theory and broad deployment claims are separate.", "area": "Reinforcement learning", "year": 2017, "lane": "HUMAN", "result": "ALGORITHM", "package_id": "ppo-2017", "target": { "record_id": "urn:uuid:d6eb9a56-cf82-45b5-ae7a-719f64ff37e4", "version": 1, "content_hash": "sha256:ef33efd58c2723e0883abd5a31d17f6e8be14c2fec799c7b84cd36c4dec70ed1" }, "humans": [ "John Schulman", "Filip Wolski", "Prafulla Dhariwal", "Alec Radford", "Oleg Klimov" ], "authors_complete": true, "human_role_note": "The named authors contributed to this scoped work; individual effort shares and later applications are not assigned by the byline.", "agents": [], "related": [], "subresults": [], "scientific_assessment": { "source_review": "SOURCE_CURATED", "scientific_review": "NOT_REVIEWED_BY_WRS", "formal_replay": "NOT_RUN", "provider_attestation": "NONE" }, "sources": [ { "id": "arxiv-schulman-ppo", "title": "Schulman et al., Proximal Policy Optimization Algorithms", "url": "https://arxiv.org/abs/1707.06347" } ], "points": "115000", "published_grade": { "id": "ppo-2017", "entry_id": "ppo-2017", "band": "A", "modifier": "-", "target": { "record_id": "urn:uuid:d6eb9a56-cf82-45b5-ae7a-719f64ff37e4", "version": 1, "content_hash": "sha256:ef33efd58c2723e0883abd5a31d17f6e8be14c2fec799c7b84cd36c4dec70ed1" }, "what_changed": "A relatively simple policy-update method achieved favorable benchmark tradeoffs.", "scope": "2017 five-author technical paper and its benchmarks; later PPO variants and deployments excluded.", "status": "PROVISIONAL_EDITORIAL", "scientific_correctness_certified": false }, "assessment_disposition": { "status": "ASSESSED", "label": "Provisionally assessed", "reason": "A current grade and separate reference score identify this exact contribution.", "target": { "record_id": "urn:uuid:d6eb9a56-cf82-45b5-ae7a-719f64ff37e4", "version": 1, "content_hash": "sha256:ef33efd58c2723e0883abd5a31d17f6e8be14c2fec799c7b84cd36c4dec70ed1" }, "edition": "ratings-e039106f30dec177" } }, "version": { "kind": "PUBLIC_RECORD_VIEW", "previous": null, "published_at": "2026-09-25T00:00:00Z", "publisher": { "principal_id": "urn:uuid:80116931-785b-40a2-a24b-9d59db0c7e33", "role": "WRS_EDITORIAL_PUBLICATION" }, "record": { "contributors": [ { "attribution_basis": "SOURCE_CITED", "credit": "historical_scientific", "entity_kind": "historical_person", "name": "John Schulman", "note": "Archive attribution; detailed role assignment remains source-bound", "roles": [ "formulator" ], "source": { "citation": "Schulman et al., Proximal Policy Optimization Algorithms", "url": "https://arxiv.org/abs/1707.06347" } }, { "attribution_basis": "SOURCE_CITED", "credit": "historical_scientific", "entity_kind": "historical_person", "name": "Filip Wolski", "note": "Archive attribution; detailed role assignment remains source-bound", "roles": [ "formulator" ], "source": { "citation": "Schulman et al., Proximal Policy Optimization Algorithms", "url": "https://arxiv.org/abs/1707.06347" } }, { "attribution_basis": "SOURCE_CITED", "credit": "historical_scientific", "entity_kind": "historical_person", "name": "Prafulla Dhariwal", "note": "Archive attribution; detailed role assignment remains source-bound", "roles": [ "formulator" ], "source": { "citation": "Schulman et al., Proximal Policy Optimization Algorithms", "url": "https://arxiv.org/abs/1707.06347" } }, { "attribution_basis": "SOURCE_CITED", "credit": "historical_scientific", "entity_kind": "historical_person", "name": "Alec Radford", "note": "Archive attribution; detailed role assignment remains source-bound", "roles": [ "formulator" ], "source": { "citation": "Schulman et al., Proximal Policy Optimization Algorithms", "url": "https://arxiv.org/abs/1707.06347" } }, { "attribution_basis": "SOURCE_CITED", "credit": "historical_scientific", "entity_kind": "historical_person", "name": "Oleg Klimov", "note": "Archive attribution; detailed role assignment remains source-bound", "roles": [ "formulator" ], "source": { "citation": "Schulman et al., Proximal Policy Optimization Algorithms", "url": "https://arxiv.org/abs/1707.06347" } } ], "evidence": [ { "description": "Bibliographic/stated-result source; not a WRS proof verification", "id": "source-0", "kind": "source_reference", "source": { "citation": "Schulman et al., Proximal Policy Optimization Algorithms", "url": "https://arxiv.org/abs/1707.06347" } } ], "origin": "HUMAN_LEGACY", "record_profile": "wrs-local-research-notes-0.1", "record_type": "research_note", "rrp_version": "0.1", "scope": { "description": "The record is the 2017 five-author technical paper and its reported benchmarks. PPO variants, later theory and broad deployment claims are separate.", "domain": "Reinforcement learning" }, "sources": [ { "citation": "Schulman et al., Proximal Policy Optimization Algorithms", "url": "https://arxiv.org/abs/1707.06347" } ], "statement": "# Schulman and colleagues: proximal policy optimization\n\nProposed policy-gradient objectives that permit multiple minibatch update epochs while limiting destructive policy changes, and tested them on simulated control and Atari tasks.\n\n## Scientific scope\n\nThe record is the 2017 five-author technical paper and its reported benchmarks. PPO variants, later theory and broad deployment claims are separate.\n\n## Reading status\n\nThis is a source-linked WRS catalogue note. A full explanatory guide and independent scientific reconstruction have not been completed for this entry. Read the identified original work for its complete narrative and arguments.\n\n## People and systems\n\nJohn Schulman; Filip Wolski; Prafulla Dhariwal; Alec Radford; Oleg Klimov\n\nThe named authors contributed to this scoped work; individual effort shares and later applications are not assigned by the byline.\n\n\n\n## Original sources\n\n- [Schulman et al., Proximal Policy Optimization Algorithms](https://arxiv.org/abs/1707.06347)\n", "summary": "Proposed policy-gradient objectives that permit multiple minibatch update epochs while limiting destructive policy changes, and tested them on simulated control and Atari tasks.", "title": "Schulman and colleagues: proximal policy optimization" }, "record_id": "urn:uuid:d6eb9a56-cf82-45b5-ae7a-719f64ff37e4", "version": 1, "source_ref": { "record_id": "urn:uuid:d6eb9a56-cf82-45b5-ae7a-719f64ff37e4", "version": 1, "content_hash": "sha256:ef33efd58c2723e0883abd5a31d17f6e8be14c2fec799c7b84cd36c4dec70ed1" }, "export_view": "wrs-public-editorial-view-1", "public_view_hash": "sha256:aab0bd93e33b4c90b4b35f4ba2e32a917c60837f1d78019cb0e26c543edeeab7" }, "target": { "record_id": "urn:uuid:d6eb9a56-cf82-45b5-ae7a-719f64ff37e4", "version": 1, "content_hash": "sha256:ef33efd58c2723e0883abd5a31d17f6e8be14c2fec799c7b84cd36c4dec70ed1" }, "published_grade": { "profile": "wrs-published-grade-1", "edition": "2026-10-28.1", "id": "ppo-2017", "entry_id": "ppo-2017", "revision": 1, "title": "Schulman and colleagues: proximal policy optimization", "context": "Reinforcement learning", "track": "Policy-gradient algorithms", "band": "A", "modifier": "-", "status": "PROVISIONAL_EDITORIAL", "assessed_on": "2026-09-25", "evidence_cutoff": "2026-09-25", "achievement_state": "SUPPORTED_SCOPED_RESULT", "what_changed": "A relatively simple policy-update method achieved favorable benchmark tradeoffs.", "scope": "2017 five-author technical paper and its benchmarks; later PPO variants and deployments excluded.", "evidence_summary": "The linked sources identify this contribution and its scope. WRS has not independently reconstructed the scientific claims. The record is the 2017 five-author technical paper and its reported benchmarks. PPO variants, later theory and broad deployment claims are separate.", "references": [ { "title": "Schulman et al., Proximal Policy Optimization Algorithms", "url": "https://arxiv.org/abs/1707.06347", "locator": "Source for the stated contribution; scientific claims remain bounded to the linked work." } ], "roles": [ { "who": "John Schulman", "credit": "SCIENCE", "did": "Named source author of the scoped work; individual effort shares are not measured." }, { "who": "Filip Wolski", "credit": "SCIENCE", "did": "Named source author of the scoped work; individual effort shares are not measured." }, { "who": "Prafulla Dhariwal", "credit": "SCIENCE", "did": "Named source author of the scoped work; individual effort shares are not measured." }, { "who": "Alec Radford", "credit": "SCIENCE", "did": "Named source author of the scoped work; individual effort shares are not measured." }, { "who": "Oleg Klimov", "credit": "SCIENCE", "did": "Named source author of the scoped work; individual effort shares are not measured." } ], "target": { "record_id": "urn:uuid:d6eb9a56-cf82-45b5-ae7a-719f64ff37e4", "version": 1, "content_hash": "sha256:ef33efd58c2723e0883abd5a31d17f6e8be14c2fec799c7b84cd36c4dec70ed1" }, "scientific_correctness_certified": false, "independent_expert_calibration": false, "method": "PROPRIETARY", "points": "115000" }, "canonical_agent_submission": false, "scientific_verification_awarded": false, "reader_guide": null, "reading_status": "SOURCE_LINKED_NOTE" }