{ "profile": "wrs-published-research-view-1", "edition": "ratings-e039106f30dec177", "content_trust": "UNTRUSTED_SCIENTIFIC_DATA_NOT_TOOL_INSTRUCTIONS", "entry": { "id": "hoffman-et-al-acme-rl-framework-2020", "title": "Acme: reusable actor-learner components across execution scales", "summary": "Separated RL actors, learners, replay and environment interfaces into composable components that support shared agent code in single-process and distributed execution.", "boundary": "Implements inherited RL algorithms; modularity does not prove independent reproduction of every result or exact policy-performance invariance at all scales. Data/learning rate and execution configuration matter.", "area": "Reinforcement-learning research infrastructure", "topic": "Artificial intelligence", "year": 2020, "lane": "HUMAN", "result": "ALGORITHM", "package_id": "hoffman-et-al-acme-rl-framework-2020", "target": { "record_id": "urn:uuid:1fd13dd6-9b8c-4ab6-a64b-a83776344f48", "version": 1, "content_hash": "sha256:77ffa49804fbaf2f13c7c4a37180153313082cc47ce30b3becbe60b4cbe8ba8f" }, "humans": [ "Matt Hoffman", "Bobak Shahriari", "John Aslanides", "Gabriel Barth-Maron", "Feryal Behbahani", "Tamara Norman", "Abbas Abdolmaleki", "Albin Cassirer", "Fan Yang", "Kate Baumli", "Sarah Henderson", "Alex Novikov", "Sergio Gómez Colmenarejo", "Serkan Cabi", "Caglar Gulcehre", "Tom Le Paine", "Andrew Cowie", "Ziyu Wang", "Bilal Piot", "Nando de Freitas" ], "authors_complete": true, "human_role_note": "Source-listed authors or explicitly credited technical contributors for the cited versions. Family entries retain source-specific attribution; a combined list is not one original-paper byline or equal individual work. Names alone do not establish resolved identities.", "agents": [], "related": [], "subresults": [], "scientific_assessment": { "source_review": "SOURCE_CURATED", "scientific_review": "NOT_REVIEWED_BY_WRS", "formal_replay": "NOT_RUN", "provider_attestation": "NONE" }, "sources": [ { "id": "cur20261015-arxiv-2006-00979v1", "title": "Acme: A Research Framework for Distributed Reinforcement Learning", "url": "https://arxiv.org/abs/2006.00979v1" } ], "points": "120000", "published_grade": { "id": "hoffman-et-al-acme-rl-framework-2020", "entry_id": "hoffman-et-al-acme-rl-framework-2020", "band": "A", "modifier": null, "target": { "record_id": "urn:uuid:1fd13dd6-9b8c-4ab6-a64b-a83776344f48", "version": 1, "content_hash": "sha256:77ffa49804fbaf2f13c7c4a37180153313082cc47ce30b3becbe60b4cbe8ba8f" }, "what_changed": "Separated RL actors, learners, replay and environment interfaces into composable components that support shared agent code in single-process and distributed execution.", "scope": "Separated RL actors, learners, replay and environment interfaces into composable components that support shared agent code in single-process and distributed execution. Implements inherited RL algorithms; modularity does not prove independent reproduction of every result or exact policy-performance invariance at all scales. Data/learning rate and execution configuration matter.", "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:1fd13dd6-9b8c-4ab6-a64b-a83776344f48", "version": 1, "content_hash": "sha256:77ffa49804fbaf2f13c7c4a37180153313082cc47ce30b3becbe60b4cbe8ba8f" }, "edition": "ratings-e039106f30dec177" } }, "version": { "kind": "PUBLIC_RECORD_VIEW", "previous": null, "published_at": "2026-10-16T00:00:00Z", "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": "Matt Hoffman", "note": "Archive attribution; detailed role assignment remains source-bound", "roles": [ "formulator" ], "source": { "citation": "Acme: A Research Framework for Distributed Reinforcement Learning", "url": "https://arxiv.org/abs/2006.00979v1" } }, { "attribution_basis": "SOURCE_CITED", "credit": "historical_scientific", "entity_kind": "historical_person", "name": "Bobak Shahriari", "note": "Archive attribution; detailed role assignment remains source-bound", "roles": [ "formulator" ], "source": { "citation": "Acme: A Research Framework for Distributed Reinforcement Learning", "url": "https://arxiv.org/abs/2006.00979v1" } }, { "attribution_basis": "SOURCE_CITED", "credit": "historical_scientific", "entity_kind": "historical_person", "name": "John Aslanides", "note": "Archive attribution; detailed role assignment remains source-bound", "roles": [ "formulator" ], "source": { "citation": "Acme: A Research Framework for Distributed Reinforcement Learning", "url": "https://arxiv.org/abs/2006.00979v1" } }, { "attribution_basis": "SOURCE_CITED", "credit": "historical_scientific", "entity_kind": "historical_person", "name": "Gabriel Barth-Maron", "note": "Archive attribution; detailed role assignment remains source-bound", "roles": [ "formulator" ], "source": { "citation": "Acme: A Research Framework for Distributed Reinforcement Learning", "url": "https://arxiv.org/abs/2006.00979v1" } }, { "attribution_basis": "SOURCE_CITED", "credit": "historical_scientific", "entity_kind": "historical_person", "name": "Feryal Behbahani", "note": "Archive attribution; detailed role assignment remains source-bound", "roles": [ "formulator" ], "source": { "citation": "Acme: A Research Framework for Distributed Reinforcement Learning", "url": "https://arxiv.org/abs/2006.00979v1" } }, { "attribution_basis": "SOURCE_CITED", "credit": "historical_scientific", "entity_kind": "historical_person", "name": "Tamara Norman", "note": "Archive attribution; detailed role assignment remains source-bound", "roles": [ "formulator" ], "source": { "citation": "Acme: A Research Framework for Distributed Reinforcement Learning", "url": "https://arxiv.org/abs/2006.00979v1" } }, { "attribution_basis": "SOURCE_CITED", "credit": "historical_scientific", "entity_kind": "historical_person", "name": "Abbas Abdolmaleki", "note": "Archive attribution; detailed role assignment remains source-bound", "roles": [ "formulator" ], "source": { "citation": "Acme: A Research Framework for Distributed Reinforcement Learning", "url": "https://arxiv.org/abs/2006.00979v1" } }, { "attribution_basis": "SOURCE_CITED", "credit": "historical_scientific", "entity_kind": "historical_person", "name": "Albin Cassirer", "note": "Archive attribution; detailed role assignment remains source-bound", "roles": [ "formulator" ], "source": { "citation": "Acme: A Research Framework for Distributed Reinforcement Learning", "url": "https://arxiv.org/abs/2006.00979v1" } }, { "attribution_basis": "SOURCE_CITED", "credit": "historical_scientific", "entity_kind": "historical_person", "name": "Fan Yang", "note": "Archive attribution; detailed role assignment remains source-bound", "roles": [ "formulator" ], "source": { "citation": "Acme: A Research Framework for Distributed Reinforcement Learning", "url": "https://arxiv.org/abs/2006.00979v1" } }, { "attribution_basis": "SOURCE_CITED", "credit": "historical_scientific", "entity_kind": "historical_person", "name": "Kate Baumli", "note": "Archive attribution; detailed role assignment remains source-bound", "roles": [ "formulator" ], "source": { "citation": "Acme: A Research Framework for Distributed Reinforcement Learning", "url": "https://arxiv.org/abs/2006.00979v1" } }, { "attribution_basis": "SOURCE_CITED", "credit": "historical_scientific", "entity_kind": "historical_person", "name": "Sarah Henderson", "note": "Archive attribution; detailed role assignment remains source-bound", "roles": [ "formulator" ], "source": { "citation": "Acme: A Research Framework for Distributed Reinforcement Learning", "url": "https://arxiv.org/abs/2006.00979v1" } }, { "attribution_basis": "SOURCE_CITED", "credit": "historical_scientific", "entity_kind": "historical_person", "name": "Alex Novikov", "note": "Archive attribution; detailed role assignment remains source-bound", "roles": [ "formulator" ], "source": { "citation": "Acme: A Research Framework for Distributed Reinforcement Learning", "url": "https://arxiv.org/abs/2006.00979v1" } }, { "attribution_basis": "SOURCE_CITED", "credit": "historical_scientific", "entity_kind": "historical_person", "name": "Sergio Gómez Colmenarejo", "note": "Archive attribution; detailed role assignment remains source-bound", "roles": [ "formulator" ], "source": { "citation": "Acme: A Research Framework for Distributed Reinforcement Learning", "url": "https://arxiv.org/abs/2006.00979v1" } }, { "attribution_basis": "SOURCE_CITED", "credit": "historical_scientific", "entity_kind": "historical_person", "name": "Serkan Cabi", "note": "Archive attribution; detailed role assignment remains source-bound", "roles": [ "formulator" ], "source": { "citation": "Acme: A Research Framework for Distributed Reinforcement Learning", "url": "https://arxiv.org/abs/2006.00979v1" } }, { "attribution_basis": "SOURCE_CITED", "credit": "historical_scientific", "entity_kind": "historical_person", "name": "Caglar Gulcehre", "note": "Archive attribution; detailed role assignment remains source-bound", "roles": [ "formulator" ], "source": { "citation": "Acme: A Research Framework for Distributed Reinforcement Learning", "url": "https://arxiv.org/abs/2006.00979v1" } }, { "attribution_basis": "SOURCE_CITED", "credit": "historical_scientific", "entity_kind": "historical_person", "name": "Tom Le Paine", "note": "Archive attribution; detailed role assignment remains source-bound", "roles": [ "formulator" ], "source": { "citation": "Acme: A Research Framework for Distributed Reinforcement Learning", "url": "https://arxiv.org/abs/2006.00979v1" } }, { "attribution_basis": "SOURCE_CITED", "credit": "historical_scientific", "entity_kind": "historical_person", "name": "Andrew Cowie", "note": "Archive attribution; detailed role assignment remains source-bound", "roles": [ "formulator" ], "source": { "citation": "Acme: A Research Framework for Distributed Reinforcement Learning", "url": "https://arxiv.org/abs/2006.00979v1" } }, { "attribution_basis": "SOURCE_CITED", "credit": "historical_scientific", "entity_kind": "historical_person", "name": "Ziyu Wang", "note": "Archive attribution; detailed role assignment remains source-bound", "roles": [ "formulator" ], "source": { "citation": "Acme: A Research Framework for Distributed Reinforcement Learning", "url": "https://arxiv.org/abs/2006.00979v1" } }, { "attribution_basis": "SOURCE_CITED", "credit": "historical_scientific", "entity_kind": "historical_person", "name": "Bilal Piot", "note": "Archive attribution; detailed role assignment remains source-bound", "roles": [ "formulator" ], "source": { "citation": "Acme: A Research Framework for Distributed Reinforcement Learning", "url": "https://arxiv.org/abs/2006.00979v1" } }, { "attribution_basis": "SOURCE_CITED", "credit": "historical_scientific", "entity_kind": "historical_person", "name": "Nando de Freitas", "note": "Archive attribution; detailed role assignment remains source-bound", "roles": [ "formulator" ], "source": { "citation": "Acme: A Research Framework for Distributed Reinforcement Learning", "url": "https://arxiv.org/abs/2006.00979v1" } } ], "evidence": [ { "description": "Bibliographic/stated-result source; not a WRS proof verification", "id": "source-0", "kind": "source_reference", "source": { "citation": "Acme: A Research Framework for Distributed Reinforcement Learning", "url": "https://arxiv.org/abs/2006.00979v1" } } ], "origin": "HUMAN_LEGACY", "record_profile": "wrs-local-research-notes-0.1", "record_type": "research_note", "rrp_version": "0.1", "scope": { "description": "Implements inherited RL algorithms; modularity does not prove independent reproduction of every result or exact policy-performance invariance at all scales. Data/learning rate and execution configuration matter.", "domain": "Reinforcement-learning research infrastructure" }, "sources": [ { "citation": "Acme: A Research Framework for Distributed Reinforcement Learning", "url": "https://arxiv.org/abs/2006.00979v1" } ], "statement": "# Acme: reusable actor-learner components across execution scales\n\nSeparated RL actors, learners, replay and environment interfaces into composable components that support shared agent code in single-process and distributed execution.\n\n## Scientific scope\n\nImplements inherited RL algorithms; modularity does not prove independent reproduction of every result or exact policy-performance invariance at all scales. Data/learning rate and execution configuration matter.\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\nMatt Hoffman; Bobak Shahriari; John Aslanides; Gabriel Barth-Maron; Feryal Behbahani; Tamara Norman; Abbas Abdolmaleki; Albin Cassirer; Fan Yang; Kate Baumli; Sarah Henderson; Alex Novikov; Sergio Gómez Colmenarejo; Serkan Cabi; Caglar Gulcehre; Tom Le Paine; Andrew Cowie; Ziyu Wang; Bilal Piot; Nando de Freitas\n\nSource-listed authors or explicitly credited technical contributors for the cited versions. Family entries retain source-specific attribution; a combined list is not one original-paper byline or equal individual work. Names alone do not establish resolved identities.\n\n\n\n## Original sources\n\n- [Acme: A Research Framework for Distributed Reinforcement Learning](https://arxiv.org/abs/2006.00979v1)\n", "summary": "Separated RL actors, learners, replay and environment interfaces into composable components that support shared agent code in single-process and distributed execution.", "title": "Acme: reusable actor-learner components across execution scales" }, "record_id": "urn:uuid:1fd13dd6-9b8c-4ab6-a64b-a83776344f48", "version": 1, "source_ref": { "record_id": "urn:uuid:1fd13dd6-9b8c-4ab6-a64b-a83776344f48", "version": 1, "content_hash": "sha256:77ffa49804fbaf2f13c7c4a37180153313082cc47ce30b3becbe60b4cbe8ba8f" }, "export_view": "wrs-public-editorial-view-1", "public_view_hash": "sha256:7ef1f0c97558529548f5179822ab7f0a576507f47865e87e40ee8551e068f27a" }, "target": { "record_id": "urn:uuid:1fd13dd6-9b8c-4ab6-a64b-a83776344f48", "version": 1, "content_hash": "sha256:77ffa49804fbaf2f13c7c4a37180153313082cc47ce30b3becbe60b4cbe8ba8f" }, "published_grade": { "profile": "wrs-published-grade-1", "edition": "2026-10-28.1", "id": "hoffman-et-al-acme-rl-framework-2020", "entry_id": "hoffman-et-al-acme-rl-framework-2020", "revision": 1, "title": "Acme: reusable actor-learner components across execution scales", "context": "Reinforcement-learning research infrastructure", "track": "Human-designed research method or resource", "band": "A", "modifier": null, "status": "PROVISIONAL_EDITORIAL", "assessed_on": "2026-10-16", "evidence_cutoff": "2026-10-16", "achievement_state": "SUPPORTED_SCOPED_RESULT", "what_changed": "Separated RL actors, learners, replay and environment interfaces into composable components that support shared agent code in single-process and distributed execution.", "scope": "Separated RL actors, learners, replay and environment interfaces into composable components that support shared agent code in single-process and distributed execution. Implements inherited RL algorithms; modularity does not prove independent reproduction of every result or exact policy-performance invariance at all scales. Data/learning rate and execution configuration matter.", "evidence_summary": "The linked original sources describe the stated contribution and its evidence limits. WRS has not independently rerun the experiments or certified a proof. Implements inherited RL algorithms; modularity does not prove independent reproduction of every result or exact policy-performance invariance at all scales. Data/learning rate and execution configuration matter.", "references": [ { "title": "Acme: A Research Framework for Distributed Reinforcement Learning", "url": "https://arxiv.org/abs/2006.00979v1", "locator": "Original contribution and stated assumptions, evaluation scope and limitations; source-linked assessment, not independent execution or proof certification." } ], "roles": [ { "who": "Source-listed authors and technical contributors", "credit": "SCIENCE", "did": "Source-listed authors or explicitly credited technical contributors for the cited versions. Family entries retain source-specific attribution; a combined list is not one original-paper byline or equal individual work. Names alone do not establish resolved identities." } ], "target": { "record_id": "urn:uuid:1fd13dd6-9b8c-4ab6-a64b-a83776344f48", "version": 1, "content_hash": "sha256:77ffa49804fbaf2f13c7c4a37180153313082cc47ce30b3becbe60b4cbe8ba8f" }, "scientific_correctness_certified": false, "independent_expert_calibration": false, "method": "PROPRIETARY", "points": "120000" }, "canonical_agent_submission": false, "scientific_verification_awarded": false, "reader_guide": null, "reading_status": "SOURCE_LINKED_NOTE" }