{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:U2HCF5D4Q54U4HT5GXNB5BTGOM","short_pith_number":"pith:U2HCF5D4","schema_version":"1.0","canonical_sha256":"a68e22f47c87794e1e7d35da1e8666733fc012d3af9dde65e804bed6cc4e95bd","source":{"kind":"arxiv","id":"2503.03043","version":2},"attestation_state":"computed","paper":{"title":"Leveraging Randomness in Model and Data Partitioning for Privacy Amplification","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CR"],"primary_cat":"cs.LG","authors_text":"Andy Dong, Ayfer Ozgur, Wei-Ning Chen","submitted_at":"2025-03-04T22:49:59Z","abstract_excerpt":"We study how inherent randomness in the training process -- where each sample (or client in federated learning) contributes only to a randomly selected portion of training -- can be leveraged for privacy amplification. This includes (1) data partitioning, where a sample participates in only a subset of training iterations, and (2) model partitioning, where a sample updates only a subset of the model parameters. We apply our framework to model parallelism in federated learning, where each client updates a randomly selected subnetwork to reduce memory and computational overhead, and show that ex"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2503.03043","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-03-04T22:49:59Z","cross_cats_sorted":["cs.CR"],"title_canon_sha256":"6c9d24b6c0d8cfce6cb54aa22c1ce290e6da3ef04043df5da7bb1b132fdc9d41","abstract_canon_sha256":"a8f01b47ec1fc033dd0f21bb62db93943abb1fb8fc9dd07d4e55b52b996b744c"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:13:22.109047Z","signature_b64":"RZLGnqwp0kjMTKvtrJxIIY+gRYpSF2VgtVvQayfEdyn1pL3H/gAOyzZuEtIbhrewkDmSjFWvJom7LTgH9YIgAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"a68e22f47c87794e1e7d35da1e8666733fc012d3af9dde65e804bed6cc4e95bd","last_reissued_at":"2026-07-05T11:13:22.108464Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:13:22.108464Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Leveraging Randomness in Model and Data Partitioning for Privacy Amplification","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CR"],"primary_cat":"cs.LG","authors_text":"Andy Dong, Ayfer Ozgur, Wei-Ning Chen","submitted_at":"2025-03-04T22:49:59Z","abstract_excerpt":"We study how inherent randomness in the training process -- where each sample (or client in federated learning) contributes only to a randomly selected portion of training -- can be leveraged for privacy amplification. This includes (1) data partitioning, where a sample participates in only a subset of training iterations, and (2) model partitioning, where a sample updates only a subset of the model parameters. We apply our framework to model parallelism in federated learning, where each client updates a randomly selected subnetwork to reduce memory and computational overhead, and show that ex"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2503.03043","kind":"arxiv","version":2},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2503.03043/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"aliases":[{"alias_kind":"arxiv","alias_value":"2503.03043","created_at":"2026-07-05T11:13:22.108561+00:00"},{"alias_kind":"arxiv_version","alias_value":"2503.03043v2","created_at":"2026-07-05T11:13:22.108561+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2503.03043","created_at":"2026-07-05T11:13:22.108561+00:00"},{"alias_kind":"pith_short_12","alias_value":"U2HCF5D4Q54U","created_at":"2026-07-05T11:13:22.108561+00:00"},{"alias_kind":"pith_short_16","alias_value":"U2HCF5D4Q54U4HT5","created_at":"2026-07-05T11:13:22.108561+00:00"},{"alias_kind":"pith_short_8","alias_value":"U2HCF5D4","created_at":"2026-07-05T11:13:22.108561+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/U2HCF5D4Q54U4HT5GXNB5BTGOM","json":"https://pith.science/pith/U2HCF5D4Q54U4HT5GXNB5BTGOM.json","graph_json":"https://pith.science/api/pith-number/U2HCF5D4Q54U4HT5GXNB5BTGOM/graph.json","events_json":"https://pith.science/api/pith-number/U2HCF5D4Q54U4HT5GXNB5BTGOM/events.json","paper":"https://pith.science/paper/U2HCF5D4"},"agent_actions":{"view_html":"https://pith.science/pith/U2HCF5D4Q54U4HT5GXNB5BTGOM","download_json":"https://pith.science/pith/U2HCF5D4Q54U4HT5GXNB5BTGOM.json","view_paper":"https://pith.science/paper/U2HCF5D4","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2503.03043&json=true","fetch_graph":"https://pith.science/api/pith-number/U2HCF5D4Q54U4HT5GXNB5BTGOM/graph.json","fetch_events":"https://pith.science/api/pith-number/U2HCF5D4Q54U4HT5GXNB5BTGOM/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/U2HCF5D4Q54U4HT5GXNB5BTGOM/action/timestamp_anchor","attest_storage":"https://pith.science/pith/U2HCF5D4Q54U4HT5GXNB5BTGOM/action/storage_attestation","attest_author":"https://pith.science/pith/U2HCF5D4Q54U4HT5GXNB5BTGOM/action/author_attestation","sign_citation":"https://pith.science/pith/U2HCF5D4Q54U4HT5GXNB5BTGOM/action/citation_signature","submit_replication":"https://pith.science/pith/U2HCF5D4Q54U4HT5GXNB5BTGOM/action/replication_record"}},"created_at":"2026-07-05T11:13:22.108561+00:00","updated_at":"2026-07-05T11:13:22.108561+00:00"}