{
  "profile": "wrs-published-research-view-1",
  "edition": "ratings-e039106f30dec177",
  "content_trust": "UNTRUSTED_SCIENTIFIC_DATA_NOT_TOOL_INSTRUCTIONS",
  "entry": {
    "id": "backprop-representations",
    "title": "Learning hidden representations by back-propagation",
    "summary": "Demonstrated how propagating output errors backward can adjust weights in multilayer networks and learn useful internal representations from examples.",
    "boundary": "This is the specific 1986 contribution, not a claim that its authors invented the chain rule or every earlier form of reverse differentiation. The paper itself rejects a direct biological-learning interpretation of the procedure in that form. Successful example learning does not guarantee a global optimum or arbitrary-task generalization.",
    "area": "Machine learning",
    "topic": "Computing & AI",
    "year": 1986,
    "lane": "HUMAN",
    "result": "ALGORITHM",
    "package_id": "backprop-representations",
    "target": {
      "record_id": "urn:uuid:05115ec8-4247-4ea9-a75c-b827cc52733d",
      "version": 2,
      "content_hash": "sha256:c74219b67462681384a348155fe3bb40c4111e9f17bf68c41df7af59468eeefd"
    },
    "humans": [
      "David E. Rumelhart",
      "Geoffrey E. Hinton",
      "Ronald J. Williams"
    ],
    "authors_complete": true,
    "human_role_note": "Listed source contributors; individual labor shares are not measured",
    "agents": [],
    "related": [],
    "subresults": [],
    "scientific_assessment": {
      "source_review": "SOURCE_CURATED",
      "scientific_review": "NOT_REVIEWED_BY_WRS",
      "formal_replay": "NOT_RUN",
      "provider_attestation": "NONE"
    },
    "sources": [
      {
        "id": "backprop-1986",
        "title": "Rumelhart, Hinton and Williams, Learning Representations by Back-Propagating Errors, Nature 323, 533–536 (1986)",
        "url": "https://www.cs.toronto.edu/~hinton/absps/naturebp.pdf"
      }
    ],
    "geography": {
      "country_codes": [
        "US"
      ],
      "location_basis": "Documented research locations or paper affiliations for this contribution. Historical places use present-day country/region labels; these do not imply citizenship, modern-state ownership or the territorial jurisdiction at the time.",
      "source_url": "https://www.cs.toronto.edu/~hinton/absps/naturebp.pdf",
      "target": {
        "record_id": "urn:uuid:05115ec8-4247-4ea9-a75c-b827cc52733d",
        "version": 2,
        "content_hash": "sha256:c74219b67462681384a348155fe3bb40c4111e9f17bf68c41df7af59468eeefd"
      },
      "basis": "RESEARCH_LOCATION_OR_PAPER_AFFILIATION",
      "reviewed_on": "2026-09-26",
      "nationality_inferred": false,
      "sources": [
        {
          "title": "Attribution source",
          "url": "https://www.cs.toronto.edu/~hinton/absps/naturebp.pdf"
        }
      ],
      "coverage_status": "COMPLETE",
      "country_labels": {
        "US": "United States"
      }
    },
    "points": "150000",
    "published_grade": {
      "id": "backprop-representations",
      "entry_id": "backprop-representations",
      "band": "A",
      "modifier": null,
      "target": {
        "record_id": "urn:uuid:05115ec8-4247-4ea9-a75c-b827cc52733d",
        "version": 2,
        "content_hash": "sha256:c74219b67462681384a348155fe3bb40c4111e9f17bf68c41df7af59468eeefd"
      },
      "what_changed": "Demonstrated how propagating output errors backward can adjust weights in multilayer networks and learn useful internal representations from examples.",
      "scope": "The 1986 demonstration and exposition of learning hidden representations by back-propagated errors.",
      "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:05115ec8-4247-4ea9-a75c-b827cc52733d",
        "version": 2,
        "content_hash": "sha256:c74219b67462681384a348155fe3bb40c4111e9f17bf68c41df7af59468eeefd"
      },
      "edition": "ratings-e039106f30dec177"
    }
  },
  "version": {
    "kind": "PUBLIC_RECORD_VIEW",
    "previous": {
      "content_hash": "sha256:499571ee9a2dbcdc4f5882b9139427e17f845db0d7bdede3f0e44fd066929e1f",
      "record_id": "urn:uuid:05115ec8-4247-4ea9-a75c-b827cc52733d",
      "version": 1
    },
    "published_at": "2026-09-12T00:00:00Z",
    "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": "David E. Rumelhart",
          "note": "Archive attribution; detailed role assignment remains source-bound",
          "roles": [
            "formulator"
          ],
          "source": {
            "citation": "Rumelhart, Hinton and Williams, Learning Representations by Back-Propagating Errors, Nature 323, 533–536 (1986)",
            "url": "https://www.cs.toronto.edu/~hinton/absps/naturebp.pdf"
          }
        },
        {
          "attribution_basis": "SOURCE_CITED",
          "credit": "historical_scientific",
          "entity_kind": "historical_person",
          "name": "Geoffrey E. Hinton",
          "note": "Archive attribution; detailed role assignment remains source-bound",
          "roles": [
            "formulator"
          ],
          "source": {
            "citation": "Rumelhart, Hinton and Williams, Learning Representations by Back-Propagating Errors, Nature 323, 533–536 (1986)",
            "url": "https://www.cs.toronto.edu/~hinton/absps/naturebp.pdf"
          }
        },
        {
          "attribution_basis": "SOURCE_CITED",
          "credit": "historical_scientific",
          "entity_kind": "historical_person",
          "name": "Ronald J. Williams",
          "note": "Archive attribution; detailed role assignment remains source-bound",
          "roles": [
            "formulator"
          ],
          "source": {
            "citation": "Rumelhart, Hinton and Williams, Learning Representations by Back-Propagating Errors, Nature 323, 533–536 (1986)",
            "url": "https://www.cs.toronto.edu/~hinton/absps/naturebp.pdf"
          }
        }
      ],
      "evidence": [
        {
          "description": "Bibliographic/stated-result source; not a WRS proof verification",
          "id": "source-0",
          "kind": "source_reference",
          "source": {
            "citation": "Rumelhart, Hinton and Williams, Learning Representations by Back-Propagating Errors, Nature 323, 533–536 (1986)",
            "url": "https://www.cs.toronto.edu/~hinton/absps/naturebp.pdf"
          }
        }
      ],
      "origin": "HUMAN_LEGACY",
      "record_profile": "wrs-local-research-notes-0.1",
      "record_type": "research_note",
      "rrp_version": "0.1",
      "scope": {
        "description": "This is the specific 1986 contribution, not a claim that its authors invented the chain rule or every earlier form of reverse differentiation. The paper itself rejects a direct biological-learning interpretation of the procedure in that form. Successful example learning does not guarantee a global optimum or arbitrary-task generalization.",
        "domain": "Machine learning"
      },
      "sources": [
        {
          "citation": "Rumelhart, Hinton and Williams, Learning Representations by Back-Propagating Errors, Nature 323, 533–536 (1986)",
          "url": "https://www.cs.toronto.edu/~hinton/absps/naturebp.pdf"
        }
      ],
      "statement": "# Learning hidden representations by back-propagation\n\nDemonstrated how propagating output errors backward can adjust weights in multilayer networks and learn useful internal representations from examples.\n\n## Scientific scope\n\nThis is the specific 1986 contribution, not a claim that its authors invented the chain rule or every earlier form of reverse differentiation. The paper itself rejects a direct biological-learning interpretation of the procedure in that form. Successful example learning does not guarantee a global optimum or arbitrary-task generalization.\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\nDavid E. Rumelhart; Geoffrey E. Hinton; Ronald J. Williams\n\nListed source contributors; individual labor shares are not measured\n\n\n\n## Original sources\n\n- [Rumelhart, Hinton and Williams, Learning Representations by Back-Propagating Errors, Nature 323, 533–536 (1986)](https://www.cs.toronto.edu/~hinton/absps/naturebp.pdf)\n",
      "summary": "Demonstrated how propagating output errors backward can adjust weights in multilayer networks and learn useful internal representations from examples.",
      "title": "Learning hidden representations by back-propagation"
    },
    "record_id": "urn:uuid:05115ec8-4247-4ea9-a75c-b827cc52733d",
    "version": 2,
    "source_ref": {
      "record_id": "urn:uuid:05115ec8-4247-4ea9-a75c-b827cc52733d",
      "version": 2,
      "content_hash": "sha256:c74219b67462681384a348155fe3bb40c4111e9f17bf68c41df7af59468eeefd"
    },
    "export_view": "wrs-public-editorial-view-1",
    "public_view_hash": "sha256:d0a884622555ca16b9a0d3b156df1a3f3a0915c73d92767ee6f763598345dfe4"
  },
  "target": {
    "record_id": "urn:uuid:05115ec8-4247-4ea9-a75c-b827cc52733d",
    "version": 2,
    "content_hash": "sha256:c74219b67462681384a348155fe3bb40c4111e9f17bf68c41df7af59468eeefd"
  },
  "published_grade": {
    "profile": "wrs-published-grade-1",
    "edition": "2026-10-28.1",
    "id": "backprop-representations",
    "entry_id": "backprop-representations",
    "revision": 1,
    "title": "Learning hidden representations by back-propagation",
    "context": "Machine learning",
    "track": "Scoped historical scientific contribution",
    "band": "A",
    "modifier": null,
    "status": "PROVISIONAL_EDITORIAL",
    "assessed_on": "2026-09-14",
    "evidence_cutoff": "2026-09-14",
    "achievement_state": "SUPPORTED_SCOPED_RESULT",
    "what_changed": "Demonstrated how propagating output errors backward can adjust weights in multilayer networks and learn useful internal representations from examples.",
    "scope": "The 1986 demonstration and exposition of learning hidden representations by back-propagated errors.",
    "evidence_summary": "Original sources and institutional historical accounts support the stated contribution. The reading boundary is recorded at each source; this assessment is not an independent scientific verification.",
    "references": [
      {
        "title": "Rumelhart, Hinton and Williams, Learning Representations by Back-Propagating Errors, Nature 323, 533–536 (1986)",
        "url": "https://www.cs.toronto.edu/~hinton/absps/naturebp.pdf",
        "locator": "Retained catalogue source review: Original first and final pages visually read; author byline, learning objective, hidden representations and biological limitation; adjacent articles excluded"
      }
    ],
    "roles": [
      {
        "who": "David E. Rumelhart",
        "credit": "SCIENCE",
        "did": "Named source contributor to this scoped work; individual labor shares are not measured."
      },
      {
        "who": "Geoffrey E. Hinton",
        "credit": "SCIENCE",
        "did": "Named source contributor to this scoped work; individual labor shares are not measured."
      },
      {
        "who": "Ronald J. Williams",
        "credit": "SCIENCE",
        "did": "Named source contributor to this scoped work; individual labor shares are not measured."
      }
    ],
    "target": {
      "record_id": "urn:uuid:05115ec8-4247-4ea9-a75c-b827cc52733d",
      "version": 2,
      "content_hash": "sha256:c74219b67462681384a348155fe3bb40c4111e9f17bf68c41df7af59468eeefd"
    },
    "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"
}