{
  "profile": "wrs-published-research-view-1",
  "edition": "ratings-e039106f30dec177",
  "content_trust": "UNTRUSTED_SCIENTIFIC_DATA_NOT_TOOL_INSTRUCTIONS",
  "entry": {
    "id": "bonawitz-et-al-federated-secure-aggregation-2016",
    "title": "Dropout-tolerant secure aggregation for federated model updates",
    "summary": "Combined pairwise pseudorandom masks, threshold secret sharing and double masking to recover sums of high-dimensional client updates despite dropouts while limiting disclosure of individual updates.",
    "boundary": "Pinned 2016 workshop protocol, not an independently checked full 2017 CCS security proof. Different server threat models require different thresholds; practical unauthenticated bootstrap assumes non-malicious key exchange and remains vulnerable to active MITM. Arbitrarily malicious colluding users/server deferred to full version. Aggregate output still reveals aggregate information and is not DP; dropout tolerance is not blanket participant-security.",
    "area": "Secure aggregation protocols",
    "topic": "Artificial intelligence",
    "year": 2016,
    "lane": "HUMAN",
    "result": "ALGORITHM",
    "package_id": "bonawitz-et-al-federated-secure-aggregation-2016",
    "target": {
      "record_id": "urn:uuid:4ba68951-0798-4e97-a451-76ef4c1c4ae5",
      "version": 1,
      "content_hash": "sha256:2f12f067efc614ac18e9f6f63ea54285082e1c5e19c1d82819c0167171d8adfe"
    },
    "humans": [
      "Keith Bonawitz",
      "Vladimir Ivanov",
      "Ben Kreuter",
      "Antonio Marcedone",
      "H. Brendan McMahan",
      "Sarvar Patel",
      "Daniel Ramage",
      "Aaron Segal",
      "Karn Seth"
    ],
    "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-1611-04482v1",
        "title": "Practical Secure Aggregation for Federated Learning on User-Held Data",
        "url": "https://arxiv.org/abs/1611.04482v1"
      }
    ],
    "points": "180000",
    "published_grade": {
      "id": "bonawitz-et-al-federated-secure-aggregation-2016",
      "entry_id": "bonawitz-et-al-federated-secure-aggregation-2016",
      "band": "A",
      "modifier": "+",
      "target": {
        "record_id": "urn:uuid:4ba68951-0798-4e97-a451-76ef4c1c4ae5",
        "version": 1,
        "content_hash": "sha256:2f12f067efc614ac18e9f6f63ea54285082e1c5e19c1d82819c0167171d8adfe"
      },
      "what_changed": "Combined pairwise pseudorandom masks, threshold secret sharing and double masking to recover sums of high-dimensional client updates despite dropouts while limiting disclosure of individual updates.",
      "scope": "Combined pairwise pseudorandom masks, threshold secret sharing and double masking to recover sums of high-dimensional client updates despite dropouts while limiting disclosure of individual updates. Pinned 2016 workshop protocol, not an independently checked full 2017 CCS security proof. Different server threat models require different thresholds; practical unauthenticated bootstrap assumes non-malicious key exchange and remains vulnerable to active MITM. Arbitrarily malicious colluding users/server deferred to full version. Aggregate output still reveals aggregate information and is not DP; dropout tolerance is not blanket participant-security.",
      "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:4ba68951-0798-4e97-a451-76ef4c1c4ae5",
        "version": 1,
        "content_hash": "sha256:2f12f067efc614ac18e9f6f63ea54285082e1c5e19c1d82819c0167171d8adfe"
      },
      "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": "Keith Bonawitz",
          "note": "Archive attribution; detailed role assignment remains source-bound",
          "roles": [
            "formulator"
          ],
          "source": {
            "citation": "Practical Secure Aggregation for Federated Learning on User-Held Data",
            "url": "https://arxiv.org/abs/1611.04482v1"
          }
        },
        {
          "attribution_basis": "SOURCE_CITED",
          "credit": "historical_scientific",
          "entity_kind": "historical_person",
          "name": "Vladimir Ivanov",
          "note": "Archive attribution; detailed role assignment remains source-bound",
          "roles": [
            "formulator"
          ],
          "source": {
            "citation": "Practical Secure Aggregation for Federated Learning on User-Held Data",
            "url": "https://arxiv.org/abs/1611.04482v1"
          }
        },
        {
          "attribution_basis": "SOURCE_CITED",
          "credit": "historical_scientific",
          "entity_kind": "historical_person",
          "name": "Ben Kreuter",
          "note": "Archive attribution; detailed role assignment remains source-bound",
          "roles": [
            "formulator"
          ],
          "source": {
            "citation": "Practical Secure Aggregation for Federated Learning on User-Held Data",
            "url": "https://arxiv.org/abs/1611.04482v1"
          }
        },
        {
          "attribution_basis": "SOURCE_CITED",
          "credit": "historical_scientific",
          "entity_kind": "historical_person",
          "name": "Antonio Marcedone",
          "note": "Archive attribution; detailed role assignment remains source-bound",
          "roles": [
            "formulator"
          ],
          "source": {
            "citation": "Practical Secure Aggregation for Federated Learning on User-Held Data",
            "url": "https://arxiv.org/abs/1611.04482v1"
          }
        },
        {
          "attribution_basis": "SOURCE_CITED",
          "credit": "historical_scientific",
          "entity_kind": "historical_person",
          "name": "H. Brendan McMahan",
          "note": "Archive attribution; detailed role assignment remains source-bound",
          "roles": [
            "formulator"
          ],
          "source": {
            "citation": "Practical Secure Aggregation for Federated Learning on User-Held Data",
            "url": "https://arxiv.org/abs/1611.04482v1"
          }
        },
        {
          "attribution_basis": "SOURCE_CITED",
          "credit": "historical_scientific",
          "entity_kind": "historical_person",
          "name": "Sarvar Patel",
          "note": "Archive attribution; detailed role assignment remains source-bound",
          "roles": [
            "formulator"
          ],
          "source": {
            "citation": "Practical Secure Aggregation for Federated Learning on User-Held Data",
            "url": "https://arxiv.org/abs/1611.04482v1"
          }
        },
        {
          "attribution_basis": "SOURCE_CITED",
          "credit": "historical_scientific",
          "entity_kind": "historical_person",
          "name": "Daniel Ramage",
          "note": "Archive attribution; detailed role assignment remains source-bound",
          "roles": [
            "formulator"
          ],
          "source": {
            "citation": "Practical Secure Aggregation for Federated Learning on User-Held Data",
            "url": "https://arxiv.org/abs/1611.04482v1"
          }
        },
        {
          "attribution_basis": "SOURCE_CITED",
          "credit": "historical_scientific",
          "entity_kind": "historical_person",
          "name": "Aaron Segal",
          "note": "Archive attribution; detailed role assignment remains source-bound",
          "roles": [
            "formulator"
          ],
          "source": {
            "citation": "Practical Secure Aggregation for Federated Learning on User-Held Data",
            "url": "https://arxiv.org/abs/1611.04482v1"
          }
        },
        {
          "attribution_basis": "SOURCE_CITED",
          "credit": "historical_scientific",
          "entity_kind": "historical_person",
          "name": "Karn Seth",
          "note": "Archive attribution; detailed role assignment remains source-bound",
          "roles": [
            "formulator"
          ],
          "source": {
            "citation": "Practical Secure Aggregation for Federated Learning on User-Held Data",
            "url": "https://arxiv.org/abs/1611.04482v1"
          }
        }
      ],
      "evidence": [
        {
          "description": "Bibliographic/stated-result source; not a WRS proof verification",
          "id": "source-0",
          "kind": "source_reference",
          "source": {
            "citation": "Practical Secure Aggregation for Federated Learning on User-Held Data",
            "url": "https://arxiv.org/abs/1611.04482v1"
          }
        }
      ],
      "origin": "HUMAN_LEGACY",
      "record_profile": "wrs-local-research-notes-0.1",
      "record_type": "research_note",
      "rrp_version": "0.1",
      "scope": {
        "description": "Pinned 2016 workshop protocol, not an independently checked full 2017 CCS security proof. Different server threat models require different thresholds; practical unauthenticated bootstrap assumes non-malicious key exchange and remains vulnerable to active MITM. Arbitrarily malicious colluding users/server deferred to full version. Aggregate output still reveals aggregate information and is not DP; dropout tolerance is not blanket participant-security.",
        "domain": "Secure aggregation protocols"
      },
      "sources": [
        {
          "citation": "Practical Secure Aggregation for Federated Learning on User-Held Data",
          "url": "https://arxiv.org/abs/1611.04482v1"
        }
      ],
      "statement": "# Dropout-tolerant secure aggregation for federated model updates\n\nCombined pairwise pseudorandom masks, threshold secret sharing and double masking to recover sums of high-dimensional client updates despite dropouts while limiting disclosure of individual updates.\n\n## Scientific scope\n\nPinned 2016 workshop protocol, not an independently checked full 2017 CCS security proof. Different server threat models require different thresholds; practical unauthenticated bootstrap assumes non-malicious key exchange and remains vulnerable to active MITM. Arbitrarily malicious colluding users/server deferred to full version. Aggregate output still reveals aggregate information and is not DP; dropout tolerance is not blanket participant-security.\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\nKeith Bonawitz; Vladimir Ivanov; Ben Kreuter; Antonio Marcedone; H. Brendan McMahan; Sarvar Patel; Daniel Ramage; Aaron Segal; Karn Seth\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- [Practical Secure Aggregation for Federated Learning on User-Held Data](https://arxiv.org/abs/1611.04482v1)\n",
      "summary": "Combined pairwise pseudorandom masks, threshold secret sharing and double masking to recover sums of high-dimensional client updates despite dropouts while limiting disclosure of individual updates.",
      "title": "Dropout-tolerant secure aggregation for federated model updates"
    },
    "record_id": "urn:uuid:4ba68951-0798-4e97-a451-76ef4c1c4ae5",
    "version": 1,
    "source_ref": {
      "record_id": "urn:uuid:4ba68951-0798-4e97-a451-76ef4c1c4ae5",
      "version": 1,
      "content_hash": "sha256:2f12f067efc614ac18e9f6f63ea54285082e1c5e19c1d82819c0167171d8adfe"
    },
    "export_view": "wrs-public-editorial-view-1",
    "public_view_hash": "sha256:ab2f7dd159b989511b0734173369a9610a742cce8e28ad37150c00006aac554c"
  },
  "target": {
    "record_id": "urn:uuid:4ba68951-0798-4e97-a451-76ef4c1c4ae5",
    "version": 1,
    "content_hash": "sha256:2f12f067efc614ac18e9f6f63ea54285082e1c5e19c1d82819c0167171d8adfe"
  },
  "published_grade": {
    "profile": "wrs-published-grade-1",
    "edition": "2026-10-28.1",
    "id": "bonawitz-et-al-federated-secure-aggregation-2016",
    "entry_id": "bonawitz-et-al-federated-secure-aggregation-2016",
    "revision": 1,
    "title": "Dropout-tolerant secure aggregation for federated model updates",
    "context": "Secure aggregation protocols",
    "track": "Human-designed research method or resource",
    "band": "A",
    "modifier": "+",
    "status": "PROVISIONAL_EDITORIAL",
    "assessed_on": "2026-10-16",
    "evidence_cutoff": "2026-10-16",
    "achievement_state": "SUPPORTED_SCOPED_RESULT",
    "what_changed": "Combined pairwise pseudorandom masks, threshold secret sharing and double masking to recover sums of high-dimensional client updates despite dropouts while limiting disclosure of individual updates.",
    "scope": "Combined pairwise pseudorandom masks, threshold secret sharing and double masking to recover sums of high-dimensional client updates despite dropouts while limiting disclosure of individual updates. Pinned 2016 workshop protocol, not an independently checked full 2017 CCS security proof. Different server threat models require different thresholds; practical unauthenticated bootstrap assumes non-malicious key exchange and remains vulnerable to active MITM. Arbitrarily malicious colluding users/server deferred to full version. Aggregate output still reveals aggregate information and is not DP; dropout tolerance is not blanket participant-security.",
    "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. Pinned 2016 workshop protocol, not an independently checked full 2017 CCS security proof. Different server threat models require different thresholds; practical unauthenticated bootstrap assumes non-malicious key exchange and remains vulnerable to active MITM. Arbitrarily malicious colluding users/server deferred to full version. Aggregate output still reveals aggregate information and is not DP; dropout tolerance is not blanket participant-security.",
    "references": [
      {
        "title": "Practical Secure Aggregation for Federated Learning on User-Held Data",
        "url": "https://arxiv.org/abs/1611.04482v1",
        "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:4ba68951-0798-4e97-a451-76ef4c1c4ae5",
      "version": 1,
      "content_hash": "sha256:2f12f067efc614ac18e9f6f63ea54285082e1c5e19c1d82819c0167171d8adfe"
    },
    "scientific_correctness_certified": false,
    "independent_expert_calibration": false,
    "method": "PROPRIETARY",
    "points": "180000"
  },
  "canonical_agent_submission": false,
  "scientific_verification_awarded": false,
  "reader_guide": null,
  "reading_status": "SOURCE_LINKED_NOTE"
}