{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:PAYCD2W2B5UXPODNXZOHJ4HGVF","short_pith_number":"pith:PAYCD2W2","schema_version":"1.0","canonical_sha256":"783021eada0f6977b86dbe5c74f0e6a94cbb7753641d03e97fd0e1ec9ec7e85e","source":{"kind":"arxiv","id":"2305.13209","version":1},"attestation_state":"computed","paper":{"title":"Faster Differentially Private Convex Optimization via Second-Order Methods","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CR","math.OC","stat.ML"],"primary_cat":"cs.LG","authors_text":"Abhradeep Thakurta, Arun Ganesh, Mahdi Haghifam, Thomas Steinke","submitted_at":"2023-05-22T16:43:36Z","abstract_excerpt":"Differentially private (stochastic) gradient descent is the workhorse of DP private machine learning in both the convex and non-convex settings. Without privacy constraints, second-order methods, like Newton's method, converge faster than first-order methods like gradient descent. In this work, we investigate the prospect of using the second-order information from the loss function to accelerate DP convex optimization. We first develop a private variant of the regularized cubic Newton method of Nesterov and Polyak, and show that for the class of strongly convex loss functions, our algorithm ha"},"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":"2305.13209","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2023-05-22T16:43:36Z","cross_cats_sorted":["cs.CR","math.OC","stat.ML"],"title_canon_sha256":"c3c3f663a403bab09c5c0bc55e68d44e076d14ca77f652d333d140f57dc57b94","abstract_canon_sha256":"0be00f55e9e3633538deaa54d7314476679c2b58a0c8b98566fa56b33d12e711"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T06:12:27.667324Z","signature_b64":"Gy2Ry2poq58Xmsy78uuQJAYIqzCV1EzI4oocZco0e1foSIDSmBGPXYKm6K2Qabf98iXyBehCiNA6BeBXFXVqDw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"783021eada0f6977b86dbe5c74f0e6a94cbb7753641d03e97fd0e1ec9ec7e85e","last_reissued_at":"2026-07-05T06:12:27.666740Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T06:12:27.666740Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Faster Differentially Private Convex Optimization via Second-Order Methods","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CR","math.OC","stat.ML"],"primary_cat":"cs.LG","authors_text":"Abhradeep Thakurta, Arun Ganesh, Mahdi Haghifam, Thomas Steinke","submitted_at":"2023-05-22T16:43:36Z","abstract_excerpt":"Differentially private (stochastic) gradient descent is the workhorse of DP private machine learning in both the convex and non-convex settings. Without privacy constraints, second-order methods, like Newton's method, converge faster than first-order methods like gradient descent. In this work, we investigate the prospect of using the second-order information from the loss function to accelerate DP convex optimization. We first develop a private variant of the regularized cubic Newton method of Nesterov and Polyak, and show that for the class of strongly convex loss functions, our algorithm ha"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2305.13209","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/2305.13209/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":"2305.13209","created_at":"2026-07-05T06:12:27.666813+00:00"},{"alias_kind":"arxiv_version","alias_value":"2305.13209v1","created_at":"2026-07-05T06:12:27.666813+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2305.13209","created_at":"2026-07-05T06:12:27.666813+00:00"},{"alias_kind":"pith_short_12","alias_value":"PAYCD2W2B5UX","created_at":"2026-07-05T06:12:27.666813+00:00"},{"alias_kind":"pith_short_16","alias_value":"PAYCD2W2B5UXPODN","created_at":"2026-07-05T06:12:27.666813+00:00"},{"alias_kind":"pith_short_8","alias_value":"PAYCD2W2","created_at":"2026-07-05T06:12:27.666813+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2506.02563","citing_title":"Privacy-Preserving Federated Convex Optimization: Balancing Partial-Participation and Efficiency via Noise Cancellation","ref_index":16,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/PAYCD2W2B5UXPODNXZOHJ4HGVF","json":"https://pith.science/pith/PAYCD2W2B5UXPODNXZOHJ4HGVF.json","graph_json":"https://pith.science/api/pith-number/PAYCD2W2B5UXPODNXZOHJ4HGVF/graph.json","events_json":"https://pith.science/api/pith-number/PAYCD2W2B5UXPODNXZOHJ4HGVF/events.json","paper":"https://pith.science/paper/PAYCD2W2"},"agent_actions":{"view_html":"https://pith.science/pith/PAYCD2W2B5UXPODNXZOHJ4HGVF","download_json":"https://pith.science/pith/PAYCD2W2B5UXPODNXZOHJ4HGVF.json","view_paper":"https://pith.science/paper/PAYCD2W2","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2305.13209&json=true","fetch_graph":"https://pith.science/api/pith-number/PAYCD2W2B5UXPODNXZOHJ4HGVF/graph.json","fetch_events":"https://pith.science/api/pith-number/PAYCD2W2B5UXPODNXZOHJ4HGVF/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/PAYCD2W2B5UXPODNXZOHJ4HGVF/action/timestamp_anchor","attest_storage":"https://pith.science/pith/PAYCD2W2B5UXPODNXZOHJ4HGVF/action/storage_attestation","attest_author":"https://pith.science/pith/PAYCD2W2B5UXPODNXZOHJ4HGVF/action/author_attestation","sign_citation":"https://pith.science/pith/PAYCD2W2B5UXPODNXZOHJ4HGVF/action/citation_signature","submit_replication":"https://pith.science/pith/PAYCD2W2B5UXPODNXZOHJ4HGVF/action/replication_record"}},"created_at":"2026-07-05T06:12:27.666813+00:00","updated_at":"2026-07-05T06:12:27.666813+00:00"}