{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:AOF4477ZEG2LCJ4VBFVEEJAXHE","short_pith_number":"pith:AOF4477Z","schema_version":"1.0","canonical_sha256":"038bce7ff921b4b12795096a4224173932896879ffba38ef04b965c89c225750","source":{"kind":"arxiv","id":"2501.14095","version":3},"attestation_state":"computed","paper":{"title":"Improved subsample-and-aggregate via the private modified winsorized mean","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"stat.ME","authors_text":"Dylan Spicker, Kelly Ramsay","submitted_at":"2025-01-23T21:03:40Z","abstract_excerpt":"We develop a univariate, differentially private mean estimator, called the private modified winsorized mean, designed to be used as the aggregator in subsample-and-aggregate. We demonstrate, via real data analysis, that common differentially private multivariate mean estimators may not perform well as the aggregator, even in large datasets, motivating our developments.We show that the modified winsorized mean is minimax optimal for several, large classes of distributions, even under adversarial contamination. We also demonstrate that, empirically, the private modified winsorized mean performs "},"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":"2501.14095","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"stat.ME","submitted_at":"2025-01-23T21:03:40Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"7f2af774051675ec7599c8ca10361f01c7596443ba43b6a3cfa36c6b797286b0","abstract_canon_sha256":"843c0dff68975855b2bdf0b5bd264823c48327c1cac21f4491ed8dd4b43346a7"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:00:04.047420Z","signature_b64":"xLGDgsT7QVvaWuoyzqgOoPnjHCJLjcES+3Id+ZwySLhcN3BgJZIg2ZqtMMQ2zdeR36fwl7rSn6cW1NoetswPDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"038bce7ff921b4b12795096a4224173932896879ffba38ef04b965c89c225750","last_reissued_at":"2026-07-05T11:00:04.046940Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:00:04.046940Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Improved subsample-and-aggregate via the private modified winsorized mean","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"stat.ME","authors_text":"Dylan Spicker, Kelly Ramsay","submitted_at":"2025-01-23T21:03:40Z","abstract_excerpt":"We develop a univariate, differentially private mean estimator, called the private modified winsorized mean, designed to be used as the aggregator in subsample-and-aggregate. We demonstrate, via real data analysis, that common differentially private multivariate mean estimators may not perform well as the aggregator, even in large datasets, motivating our developments.We show that the modified winsorized mean is minimax optimal for several, large classes of distributions, even under adversarial contamination. We also demonstrate that, empirically, the private modified winsorized mean performs "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2501.14095","kind":"arxiv","version":3},"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/2501.14095/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":"2501.14095","created_at":"2026-07-05T11:00:04.047001+00:00"},{"alias_kind":"arxiv_version","alias_value":"2501.14095v3","created_at":"2026-07-05T11:00:04.047001+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2501.14095","created_at":"2026-07-05T11:00:04.047001+00:00"},{"alias_kind":"pith_short_12","alias_value":"AOF4477ZEG2L","created_at":"2026-07-05T11:00:04.047001+00:00"},{"alias_kind":"pith_short_16","alias_value":"AOF4477ZEG2LCJ4V","created_at":"2026-07-05T11:00:04.047001+00:00"},{"alias_kind":"pith_short_8","alias_value":"AOF4477Z","created_at":"2026-07-05T11:00:04.047001+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.12654","citing_title":"Computationally tractable robust differentially private mean estimation","ref_index":27,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/AOF4477ZEG2LCJ4VBFVEEJAXHE","json":"https://pith.science/pith/AOF4477ZEG2LCJ4VBFVEEJAXHE.json","graph_json":"https://pith.science/api/pith-number/AOF4477ZEG2LCJ4VBFVEEJAXHE/graph.json","events_json":"https://pith.science/api/pith-number/AOF4477ZEG2LCJ4VBFVEEJAXHE/events.json","paper":"https://pith.science/paper/AOF4477Z"},"agent_actions":{"view_html":"https://pith.science/pith/AOF4477ZEG2LCJ4VBFVEEJAXHE","download_json":"https://pith.science/pith/AOF4477ZEG2LCJ4VBFVEEJAXHE.json","view_paper":"https://pith.science/paper/AOF4477Z","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2501.14095&json=true","fetch_graph":"https://pith.science/api/pith-number/AOF4477ZEG2LCJ4VBFVEEJAXHE/graph.json","fetch_events":"https://pith.science/api/pith-number/AOF4477ZEG2LCJ4VBFVEEJAXHE/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/AOF4477ZEG2LCJ4VBFVEEJAXHE/action/timestamp_anchor","attest_storage":"https://pith.science/pith/AOF4477ZEG2LCJ4VBFVEEJAXHE/action/storage_attestation","attest_author":"https://pith.science/pith/AOF4477ZEG2LCJ4VBFVEEJAXHE/action/author_attestation","sign_citation":"https://pith.science/pith/AOF4477ZEG2LCJ4VBFVEEJAXHE/action/citation_signature","submit_replication":"https://pith.science/pith/AOF4477ZEG2LCJ4VBFVEEJAXHE/action/replication_record"}},"created_at":"2026-07-05T11:00:04.047001+00:00","updated_at":"2026-07-05T11:00:04.047001+00:00"}