{
  "profile": "wrs-published-research-view-1",
  "edition": "ratings-e039106f30dec177",
  "content_trust": "UNTRUSTED_SCIENTIFIC_DATA_NOT_TOOL_INSTRUCTIONS",
  "entry": {
    "id": "jaderberg-et-al-ftw-capture-flag-2018",
    "title": "For The Win: population-trained visual team play",
    "summary": "Combined concurrent populations, two-tier optimization of internal reward schemes and multiscale recurrent representations to learn pixel-based QuakeIII capture-the-flag policies that cooperated and competed with unfamiliar teammates/opponents.",
    "boundary": "Original2018v1 study, not a silent merge with expanded2019journal scenarios. Uses game-point event signals and evolved reward shaping, not pixels plus terminal win alone. Human/agent control resolution, reaction time and tagging accuracy differ; artificial accuracy reduction tests do not prove identical interfaces or universal human-level teamwork. PBT and earlier recurrent/RL ideas are inherited.",
    "area": "Multiagent game-learning systems",
    "topic": "Artificial intelligence",
    "year": 2018,
    "lane": "HUMAN",
    "result": "EMPIRICAL",
    "package_id": "jaderberg-et-al-ftw-capture-flag-2018",
    "target": {
      "record_id": "urn:uuid:d51081c3-d0d0-478f-affb-873067945d0f",
      "version": 1,
      "content_hash": "sha256:4ae91e5fb1cfb3201f6e02fb66344f9b963529ee3e917fc125be53e6fc8bc3be"
    },
    "humans": [
      "Max Jaderberg",
      "Wojciech M. Czarnecki",
      "Iain Dunning",
      "Luke Marris",
      "Guy Lever",
      "Antonio Garcia Castaneda",
      "Charles Beattie",
      "Neil C. Rabinowitz",
      "Ari S. Morcos",
      "Avraham Ruderman",
      "Nicolas Sonnerat",
      "Tim Green",
      "Louise Deason",
      "Joel Z. Leibo",
      "David Silver",
      "Demis Hassabis",
      "Koray Kavukcuoglu",
      "Thore Graepel"
    ],
    "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-1807-01281v1",
        "title": "Human-level performance in first-person multiplayer games with population-based deep reinforcement learning",
        "url": "https://arxiv.org/pdf/1807.01281v1"
      }
    ],
    "points": "150000",
    "published_grade": {
      "id": "jaderberg-et-al-ftw-capture-flag-2018",
      "entry_id": "jaderberg-et-al-ftw-capture-flag-2018",
      "band": "A",
      "modifier": null,
      "target": {
        "record_id": "urn:uuid:d51081c3-d0d0-478f-affb-873067945d0f",
        "version": 1,
        "content_hash": "sha256:4ae91e5fb1cfb3201f6e02fb66344f9b963529ee3e917fc125be53e6fc8bc3be"
      },
      "what_changed": "Combined concurrent populations, two-tier optimization of internal reward schemes and multiscale recurrent representations to learn pixel-based QuakeIII capture-the-flag policies that cooperated and competed with unfamiliar teammates/opponents.",
      "scope": "Combined concurrent populations, two-tier optimization of internal reward schemes and multiscale recurrent representations to learn pixel-based QuakeIII capture-the-flag policies that cooperated and competed with unfamiliar teammates/opponents. Original2018v1 study, not a silent merge with expanded2019journal scenarios. Uses game-point event signals and evolved reward shaping, not pixels plus terminal win alone. Human/agent control resolution, reaction time and tagging accuracy differ; artificial accuracy reduction tests do not prove identical interfaces or universal human-level teamwork. PBT and earlier recurrent/RL ideas are inherited.",
      "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:d51081c3-d0d0-478f-affb-873067945d0f",
        "version": 1,
        "content_hash": "sha256:4ae91e5fb1cfb3201f6e02fb66344f9b963529ee3e917fc125be53e6fc8bc3be"
      },
      "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": "Max Jaderberg",
          "note": "Archive attribution; detailed role assignment remains source-bound",
          "roles": [
            "discoverer"
          ],
          "source": {
            "citation": "Human-level performance in first-person multiplayer games with population-based deep reinforcement learning",
            "url": "https://arxiv.org/pdf/1807.01281v1"
          }
        },
        {
          "attribution_basis": "SOURCE_CITED",
          "credit": "historical_scientific",
          "entity_kind": "historical_person",
          "name": "Wojciech M. Czarnecki",
          "note": "Archive attribution; detailed role assignment remains source-bound",
          "roles": [
            "discoverer"
          ],
          "source": {
            "citation": "Human-level performance in first-person multiplayer games with population-based deep reinforcement learning",
            "url": "https://arxiv.org/pdf/1807.01281v1"
          }
        },
        {
          "attribution_basis": "SOURCE_CITED",
          "credit": "historical_scientific",
          "entity_kind": "historical_person",
          "name": "Iain Dunning",
          "note": "Archive attribution; detailed role assignment remains source-bound",
          "roles": [
            "discoverer"
          ],
          "source": {
            "citation": "Human-level performance in first-person multiplayer games with population-based deep reinforcement learning",
            "url": "https://arxiv.org/pdf/1807.01281v1"
          }
        },
        {
          "attribution_basis": "SOURCE_CITED",
          "credit": "historical_scientific",
          "entity_kind": "historical_person",
          "name": "Luke Marris",
          "note": "Archive attribution; detailed role assignment remains source-bound",
          "roles": [
            "discoverer"
          ],
          "source": {
            "citation": "Human-level performance in first-person multiplayer games with population-based deep reinforcement learning",
            "url": "https://arxiv.org/pdf/1807.01281v1"
          }
        },
        {
          "attribution_basis": "SOURCE_CITED",
          "credit": "historical_scientific",
          "entity_kind": "historical_person",
          "name": "Guy Lever",
          "note": "Archive attribution; detailed role assignment remains source-bound",
          "roles": [
            "discoverer"
          ],
          "source": {
            "citation": "Human-level performance in first-person multiplayer games with population-based deep reinforcement learning",
            "url": "https://arxiv.org/pdf/1807.01281v1"
          }
        },
        {
          "attribution_basis": "SOURCE_CITED",
          "credit": "historical_scientific",
          "entity_kind": "historical_person",
          "name": "Antonio Garcia Castaneda",
          "note": "Archive attribution; detailed role assignment remains source-bound",
          "roles": [
            "discoverer"
          ],
          "source": {
            "citation": "Human-level performance in first-person multiplayer games with population-based deep reinforcement learning",
            "url": "https://arxiv.org/pdf/1807.01281v1"
          }
        },
        {
          "attribution_basis": "SOURCE_CITED",
          "credit": "historical_scientific",
          "entity_kind": "historical_person",
          "name": "Charles Beattie",
          "note": "Archive attribution; detailed role assignment remains source-bound",
          "roles": [
            "discoverer"
          ],
          "source": {
            "citation": "Human-level performance in first-person multiplayer games with population-based deep reinforcement learning",
            "url": "https://arxiv.org/pdf/1807.01281v1"
          }
        },
        {
          "attribution_basis": "SOURCE_CITED",
          "credit": "historical_scientific",
          "entity_kind": "historical_person",
          "name": "Neil C. Rabinowitz",
          "note": "Archive attribution; detailed role assignment remains source-bound",
          "roles": [
            "discoverer"
          ],
          "source": {
            "citation": "Human-level performance in first-person multiplayer games with population-based deep reinforcement learning",
            "url": "https://arxiv.org/pdf/1807.01281v1"
          }
        },
        {
          "attribution_basis": "SOURCE_CITED",
          "credit": "historical_scientific",
          "entity_kind": "historical_person",
          "name": "Ari S. Morcos",
          "note": "Archive attribution; detailed role assignment remains source-bound",
          "roles": [
            "discoverer"
          ],
          "source": {
            "citation": "Human-level performance in first-person multiplayer games with population-based deep reinforcement learning",
            "url": "https://arxiv.org/pdf/1807.01281v1"
          }
        },
        {
          "attribution_basis": "SOURCE_CITED",
          "credit": "historical_scientific",
          "entity_kind": "historical_person",
          "name": "Avraham Ruderman",
          "note": "Archive attribution; detailed role assignment remains source-bound",
          "roles": [
            "discoverer"
          ],
          "source": {
            "citation": "Human-level performance in first-person multiplayer games with population-based deep reinforcement learning",
            "url": "https://arxiv.org/pdf/1807.01281v1"
          }
        },
        {
          "attribution_basis": "SOURCE_CITED",
          "credit": "historical_scientific",
          "entity_kind": "historical_person",
          "name": "Nicolas Sonnerat",
          "note": "Archive attribution; detailed role assignment remains source-bound",
          "roles": [
            "discoverer"
          ],
          "source": {
            "citation": "Human-level performance in first-person multiplayer games with population-based deep reinforcement learning",
            "url": "https://arxiv.org/pdf/1807.01281v1"
          }
        },
        {
          "attribution_basis": "SOURCE_CITED",
          "credit": "historical_scientific",
          "entity_kind": "historical_person",
          "name": "Tim Green",
          "note": "Archive attribution; detailed role assignment remains source-bound",
          "roles": [
            "discoverer"
          ],
          "source": {
            "citation": "Human-level performance in first-person multiplayer games with population-based deep reinforcement learning",
            "url": "https://arxiv.org/pdf/1807.01281v1"
          }
        },
        {
          "attribution_basis": "SOURCE_CITED",
          "credit": "historical_scientific",
          "entity_kind": "historical_person",
          "name": "Louise Deason",
          "note": "Archive attribution; detailed role assignment remains source-bound",
          "roles": [
            "discoverer"
          ],
          "source": {
            "citation": "Human-level performance in first-person multiplayer games with population-based deep reinforcement learning",
            "url": "https://arxiv.org/pdf/1807.01281v1"
          }
        },
        {
          "attribution_basis": "SOURCE_CITED",
          "credit": "historical_scientific",
          "entity_kind": "historical_person",
          "name": "Joel Z. Leibo",
          "note": "Archive attribution; detailed role assignment remains source-bound",
          "roles": [
            "discoverer"
          ],
          "source": {
            "citation": "Human-level performance in first-person multiplayer games with population-based deep reinforcement learning",
            "url": "https://arxiv.org/pdf/1807.01281v1"
          }
        },
        {
          "attribution_basis": "SOURCE_CITED",
          "credit": "historical_scientific",
          "entity_kind": "historical_person",
          "name": "David Silver",
          "note": "Archive attribution; detailed role assignment remains source-bound",
          "roles": [
            "discoverer"
          ],
          "source": {
            "citation": "Human-level performance in first-person multiplayer games with population-based deep reinforcement learning",
            "url": "https://arxiv.org/pdf/1807.01281v1"
          }
        },
        {
          "attribution_basis": "SOURCE_CITED",
          "credit": "historical_scientific",
          "entity_kind": "historical_person",
          "name": "Demis Hassabis",
          "note": "Archive attribution; detailed role assignment remains source-bound",
          "roles": [
            "discoverer"
          ],
          "source": {
            "citation": "Human-level performance in first-person multiplayer games with population-based deep reinforcement learning",
            "url": "https://arxiv.org/pdf/1807.01281v1"
          }
        },
        {
          "attribution_basis": "SOURCE_CITED",
          "credit": "historical_scientific",
          "entity_kind": "historical_person",
          "name": "Koray Kavukcuoglu",
          "note": "Archive attribution; detailed role assignment remains source-bound",
          "roles": [
            "discoverer"
          ],
          "source": {
            "citation": "Human-level performance in first-person multiplayer games with population-based deep reinforcement learning",
            "url": "https://arxiv.org/pdf/1807.01281v1"
          }
        },
        {
          "attribution_basis": "SOURCE_CITED",
          "credit": "historical_scientific",
          "entity_kind": "historical_person",
          "name": "Thore Graepel",
          "note": "Archive attribution; detailed role assignment remains source-bound",
          "roles": [
            "discoverer"
          ],
          "source": {
            "citation": "Human-level performance in first-person multiplayer games with population-based deep reinforcement learning",
            "url": "https://arxiv.org/pdf/1807.01281v1"
          }
        }
      ],
      "evidence": [
        {
          "description": "Bibliographic/stated-result source; not a WRS proof verification",
          "id": "source-0",
          "kind": "source_reference",
          "source": {
            "citation": "Human-level performance in first-person multiplayer games with population-based deep reinforcement learning",
            "url": "https://arxiv.org/pdf/1807.01281v1"
          }
        }
      ],
      "origin": "HUMAN_LEGACY",
      "record_profile": "wrs-local-research-notes-0.1",
      "record_type": "research_note",
      "rrp_version": "0.1",
      "scope": {
        "description": "Original2018v1 study, not a silent merge with expanded2019journal scenarios. Uses game-point event signals and evolved reward shaping, not pixels plus terminal win alone. Human/agent control resolution, reaction time and tagging accuracy differ; artificial accuracy reduction tests do not prove identical interfaces or universal human-level teamwork. PBT and earlier recurrent/RL ideas are inherited.",
        "domain": "Multiagent game-learning systems"
      },
      "sources": [
        {
          "citation": "Human-level performance in first-person multiplayer games with population-based deep reinforcement learning",
          "url": "https://arxiv.org/pdf/1807.01281v1"
        }
      ],
      "statement": "# For The Win: population-trained visual team play\n\nCombined concurrent populations, two-tier optimization of internal reward schemes and multiscale recurrent representations to learn pixel-based QuakeIII capture-the-flag policies that cooperated and competed with unfamiliar teammates/opponents.\n\n## Scientific scope\n\nOriginal2018v1 study, not a silent merge with expanded2019journal scenarios. Uses game-point event signals and evolved reward shaping, not pixels plus terminal win alone. Human/agent control resolution, reaction time and tagging accuracy differ; artificial accuracy reduction tests do not prove identical interfaces or universal human-level teamwork. PBT and earlier recurrent/RL ideas are inherited.\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\nMax Jaderberg; Wojciech M. Czarnecki; Iain Dunning; Luke Marris; Guy Lever; Antonio Garcia Castaneda; Charles Beattie; Neil C. Rabinowitz; Ari S. Morcos; Avraham Ruderman; Nicolas Sonnerat; Tim Green; Louise Deason; Joel Z. Leibo; David Silver; Demis Hassabis; Koray Kavukcuoglu; Thore Graepel\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- [Human-level performance in first-person multiplayer games with population-based deep reinforcement learning](https://arxiv.org/pdf/1807.01281v1)\n",
      "summary": "Combined concurrent populations, two-tier optimization of internal reward schemes and multiscale recurrent representations to learn pixel-based QuakeIII capture-the-flag policies that cooperated and competed with unfamiliar teammates/opponents.",
      "title": "For The Win: population-trained visual team play"
    },
    "record_id": "urn:uuid:d51081c3-d0d0-478f-affb-873067945d0f",
    "version": 1,
    "source_ref": {
      "record_id": "urn:uuid:d51081c3-d0d0-478f-affb-873067945d0f",
      "version": 1,
      "content_hash": "sha256:4ae91e5fb1cfb3201f6e02fb66344f9b963529ee3e917fc125be53e6fc8bc3be"
    },
    "export_view": "wrs-public-editorial-view-1",
    "public_view_hash": "sha256:0efa080903bd12d17ec3a9d1443bdebde7d9f20f255d52d79a8ac755eed60bf8"
  },
  "target": {
    "record_id": "urn:uuid:d51081c3-d0d0-478f-affb-873067945d0f",
    "version": 1,
    "content_hash": "sha256:4ae91e5fb1cfb3201f6e02fb66344f9b963529ee3e917fc125be53e6fc8bc3be"
  },
  "published_grade": {
    "profile": "wrs-published-grade-1",
    "edition": "2026-10-28.1",
    "id": "jaderberg-et-al-ftw-capture-flag-2018",
    "entry_id": "jaderberg-et-al-ftw-capture-flag-2018",
    "revision": 1,
    "title": "For The Win: population-trained visual team play",
    "context": "Multiagent game-learning systems",
    "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": "Combined concurrent populations, two-tier optimization of internal reward schemes and multiscale recurrent representations to learn pixel-based QuakeIII capture-the-flag policies that cooperated and competed with unfamiliar teammates/opponents.",
    "scope": "Combined concurrent populations, two-tier optimization of internal reward schemes and multiscale recurrent representations to learn pixel-based QuakeIII capture-the-flag policies that cooperated and competed with unfamiliar teammates/opponents. Original2018v1 study, not a silent merge with expanded2019journal scenarios. Uses game-point event signals and evolved reward shaping, not pixels plus terminal win alone. Human/agent control resolution, reaction time and tagging accuracy differ; artificial accuracy reduction tests do not prove identical interfaces or universal human-level teamwork. PBT and earlier recurrent/RL ideas are inherited.",
    "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. Original2018v1 study, not a silent merge with expanded2019journal scenarios. Uses game-point event signals and evolved reward shaping, not pixels plus terminal win alone. Human/agent control resolution, reaction time and tagging accuracy differ; artificial accuracy reduction tests do not prove identical interfaces or universal human-level teamwork. PBT and earlier recurrent/RL ideas are inherited.",
    "references": [
      {
        "title": "Human-level performance in first-person multiplayer games with population-based deep reinforcement learning",
        "url": "https://arxiv.org/pdf/1807.01281v1",
        "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:d51081c3-d0d0-478f-affb-873067945d0f",
      "version": 1,
      "content_hash": "sha256:4ae91e5fb1cfb3201f6e02fb66344f9b963529ee3e917fc125be53e6fc8bc3be"
    },
    "scientific_correctness_certified": false,
    "independent_expert_calibration": false,
    "method": "PROPRIETARY",
    "points": "150000"
  },
  "canonical_agent_submission": false,
  "scientific_verification_awarded": false,
  "reader_guide": null,
  "reading_status": "SOURCE_LINKED_NOTE"
}