{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:BRLGXFJFGFZOZ5GPNI745TNOF2","short_pith_number":"pith:BRLGXFJF","schema_version":"1.0","canonical_sha256":"0c566b95253172ecf4cf6a3fcecdae2e8f10e31447f2a4fbdddfbb0b1afe78f4","source":{"kind":"arxiv","id":"2502.13482","version":1},"attestation_state":"computed","paper":{"title":"Smoothed Normalization for Efficient Distributed Private Optimization","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CR","cs.DC","math.OC","stat.ML"],"primary_cat":"cs.LG","authors_text":"Egor Shulgin, Peter Richt\\'arik, Sarit Khirirat","submitted_at":"2025-02-19T07:10:32Z","abstract_excerpt":"Federated learning enables training machine learning models while preserving the privacy of participants. Surprisingly, there is no differentially private distributed method for smooth, non-convex optimization problems. The reason is that standard privacy techniques require bounding the participants' contributions, usually enforced via $\\textit{clipping}$ of the updates. Existing literature typically ignores the effect of clipping by assuming the boundedness of gradient norms or analyzes distributed algorithms with clipping but ignores DP constraints. In this work, we study an alternative appr"},"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":"2502.13482","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-02-19T07:10:32Z","cross_cats_sorted":["cs.CR","cs.DC","math.OC","stat.ML"],"title_canon_sha256":"a436c43169d69bbc29692266ebca0ad5c98cc6164d0f6bc493839ea92f325e18","abstract_canon_sha256":"4693f1134c6ed37401c163ba422b07cf6be96053992edb4105a1964d4caff64f"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:16:47.919085Z","signature_b64":"cx2oyrOeqUN3fynmCGyBgwc62xk8td6T+21nXoOQWDkakKw8Cmunu/foe6nmqtdW7hxVfP/UtsOD2uzllBQGAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"0c566b95253172ecf4cf6a3fcecdae2e8f10e31447f2a4fbdddfbb0b1afe78f4","last_reissued_at":"2026-07-05T10:16:47.918626Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:16:47.918626Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Smoothed Normalization for Efficient Distributed Private Optimization","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CR","cs.DC","math.OC","stat.ML"],"primary_cat":"cs.LG","authors_text":"Egor Shulgin, Peter Richt\\'arik, Sarit Khirirat","submitted_at":"2025-02-19T07:10:32Z","abstract_excerpt":"Federated learning enables training machine learning models while preserving the privacy of participants. Surprisingly, there is no differentially private distributed method for smooth, non-convex optimization problems. The reason is that standard privacy techniques require bounding the participants' contributions, usually enforced via $\\textit{clipping}$ of the updates. Existing literature typically ignores the effect of clipping by assuming the boundedness of gradient norms or analyzes distributed algorithms with clipping but ignores DP constraints. In this work, we study an alternative appr"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2502.13482","kind":"arxiv","version":1},"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/2502.13482/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":"2502.13482","created_at":"2026-07-05T10:16:47.918684+00:00"},{"alias_kind":"arxiv_version","alias_value":"2502.13482v1","created_at":"2026-07-05T10:16:47.918684+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2502.13482","created_at":"2026-07-05T10:16:47.918684+00:00"},{"alias_kind":"pith_short_12","alias_value":"BRLGXFJFGFZO","created_at":"2026-07-05T10:16:47.918684+00:00"},{"alias_kind":"pith_short_16","alias_value":"BRLGXFJFGFZOZ5GP","created_at":"2026-07-05T10:16:47.918684+00:00"},{"alias_kind":"pith_short_8","alias_value":"BRLGXFJF","created_at":"2026-07-05T10:16:47.918684+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2505.03736","citing_title":"Decentralized Nonconvex Optimization under Heavy-Tailed Noise: Normalization and Optimal Convergence","ref_index":62,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/BRLGXFJFGFZOZ5GPNI745TNOF2","json":"https://pith.science/pith/BRLGXFJFGFZOZ5GPNI745TNOF2.json","graph_json":"https://pith.science/api/pith-number/BRLGXFJFGFZOZ5GPNI745TNOF2/graph.json","events_json":"https://pith.science/api/pith-number/BRLGXFJFGFZOZ5GPNI745TNOF2/events.json","paper":"https://pith.science/paper/BRLGXFJF"},"agent_actions":{"view_html":"https://pith.science/pith/BRLGXFJFGFZOZ5GPNI745TNOF2","download_json":"https://pith.science/pith/BRLGXFJFGFZOZ5GPNI745TNOF2.json","view_paper":"https://pith.science/paper/BRLGXFJF","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2502.13482&json=true","fetch_graph":"https://pith.science/api/pith-number/BRLGXFJFGFZOZ5GPNI745TNOF2/graph.json","fetch_events":"https://pith.science/api/pith-number/BRLGXFJFGFZOZ5GPNI745TNOF2/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/BRLGXFJFGFZOZ5GPNI745TNOF2/action/timestamp_anchor","attest_storage":"https://pith.science/pith/BRLGXFJFGFZOZ5GPNI745TNOF2/action/storage_attestation","attest_author":"https://pith.science/pith/BRLGXFJFGFZOZ5GPNI745TNOF2/action/author_attestation","sign_citation":"https://pith.science/pith/BRLGXFJFGFZOZ5GPNI745TNOF2/action/citation_signature","submit_replication":"https://pith.science/pith/BRLGXFJFGFZOZ5GPNI745TNOF2/action/replication_record"}},"created_at":"2026-07-05T10:16:47.918684+00:00","updated_at":"2026-07-05T10:16:47.918684+00:00"}