{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:FMMWMS6ORZACUTE6WHANVJEX6B","short_pith_number":"pith:FMMWMS6O","schema_version":"1.0","canonical_sha256":"2b19664bce8e402a4c9eb1c0daa497f07d10058a0ad4b8881deb9b9073c51645","source":{"kind":"arxiv","id":"2102.04704","version":2},"attestation_state":"computed","paper":{"title":"Output Perturbation for Differentially Private Convex Optimization: Faster and More General","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CR","stat.ML"],"primary_cat":"cs.LG","authors_text":"Andrew Lowy, Meisam Razaviyayn","submitted_at":"2021-02-09T08:47:06Z","abstract_excerpt":"Finding efficient, easily implementable differentially private (DP) algorithms that offer strong excess risk bounds is an important problem in modern machine learning. To date, most work has focused on private empirical risk minimization (ERM) or private stochastic convex optimization (SCO), which corresponds to population loss minimization. However, there are often other objectives-such as fairness, adversarial robustness, or sensitivity to outliers-besides average performance that are not captured in the classical ERM/SCO setups. Further, most recent work in private SCO has focused on $(\\var"},"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":"2102.04704","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2021-02-09T08:47:06Z","cross_cats_sorted":["cs.CR","stat.ML"],"title_canon_sha256":"70ad74488c58b12b416f970407bf24165a545e5f18351fdef5c1b517876ebf71","abstract_canon_sha256":"c6318edb6e07d9887b676e7810cdba81eee668dbb1f3626017ff0d0e8c78c6b1"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:09:16.057090Z","signature_b64":"y5pZXAuaj8Cc13GH8xEWJHda1mi8dRig8ZTpv0aHHuAZ2SVSsqqdtGk0BdcJi7F31s9iuPmCBn8VjidyFCb3DQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"2b19664bce8e402a4c9eb1c0daa497f07d10058a0ad4b8881deb9b9073c51645","last_reissued_at":"2026-07-05T09:09:16.056577Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:09:16.056577Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Output Perturbation for Differentially Private Convex Optimization: Faster and More General","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CR","stat.ML"],"primary_cat":"cs.LG","authors_text":"Andrew Lowy, Meisam Razaviyayn","submitted_at":"2021-02-09T08:47:06Z","abstract_excerpt":"Finding efficient, easily implementable differentially private (DP) algorithms that offer strong excess risk bounds is an important problem in modern machine learning. To date, most work has focused on private empirical risk minimization (ERM) or private stochastic convex optimization (SCO), which corresponds to population loss minimization. However, there are often other objectives-such as fairness, adversarial robustness, or sensitivity to outliers-besides average performance that are not captured in the classical ERM/SCO setups. Further, most recent work in private SCO has focused on $(\\var"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2102.04704","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/2102.04704/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":"2102.04704","created_at":"2026-07-05T09:09:16.056635+00:00"},{"alias_kind":"arxiv_version","alias_value":"2102.04704v2","created_at":"2026-07-05T09:09:16.056635+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2102.04704","created_at":"2026-07-05T09:09:16.056635+00:00"},{"alias_kind":"pith_short_12","alias_value":"FMMWMS6ORZAC","created_at":"2026-07-05T09:09:16.056635+00:00"},{"alias_kind":"pith_short_16","alias_value":"FMMWMS6ORZACUTE6","created_at":"2026-07-05T09:09:16.056635+00:00"},{"alias_kind":"pith_short_8","alias_value":"FMMWMS6O","created_at":"2026-07-05T09:09:16.056635+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":3,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.01527","citing_title":"Near-Optimal Pure Machine Unlearning for Smooth Strongly Convex Losses","ref_index":187,"is_internal_anchor":false},{"citing_arxiv_id":"2604.15851","citing_title":"DPrivBench: Benchmarking LLMs' Reasoning for Differential Privacy","ref_index":21,"is_internal_anchor":false},{"citing_arxiv_id":"2604.15851","citing_title":"DPrivBench: Benchmarking LLMs' Reasoning for Differential Privacy","ref_index":21,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/FMMWMS6ORZACUTE6WHANVJEX6B","json":"https://pith.science/pith/FMMWMS6ORZACUTE6WHANVJEX6B.json","graph_json":"https://pith.science/api/pith-number/FMMWMS6ORZACUTE6WHANVJEX6B/graph.json","events_json":"https://pith.science/api/pith-number/FMMWMS6ORZACUTE6WHANVJEX6B/events.json","paper":"https://pith.science/paper/FMMWMS6O"},"agent_actions":{"view_html":"https://pith.science/pith/FMMWMS6ORZACUTE6WHANVJEX6B","download_json":"https://pith.science/pith/FMMWMS6ORZACUTE6WHANVJEX6B.json","view_paper":"https://pith.science/paper/FMMWMS6O","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2102.04704&json=true","fetch_graph":"https://pith.science/api/pith-number/FMMWMS6ORZACUTE6WHANVJEX6B/graph.json","fetch_events":"https://pith.science/api/pith-number/FMMWMS6ORZACUTE6WHANVJEX6B/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/FMMWMS6ORZACUTE6WHANVJEX6B/action/timestamp_anchor","attest_storage":"https://pith.science/pith/FMMWMS6ORZACUTE6WHANVJEX6B/action/storage_attestation","attest_author":"https://pith.science/pith/FMMWMS6ORZACUTE6WHANVJEX6B/action/author_attestation","sign_citation":"https://pith.science/pith/FMMWMS6ORZACUTE6WHANVJEX6B/action/citation_signature","submit_replication":"https://pith.science/pith/FMMWMS6ORZACUTE6WHANVJEX6B/action/replication_record"}},"created_at":"2026-07-05T09:09:16.056635+00:00","updated_at":"2026-07-05T09:09:16.056635+00:00"}