{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2022:ZICGYLA2UWRQ2YHZ2C5J7GFFCA","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"8e05f06c6ff4bffd189f844c3ded8f1caf9282e5e402b767cbfed8894024ac6a","cross_cats_sorted":["cs.LG"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CR","submitted_at":"2022-01-26T09:44:13Z","title_canon_sha256":"1d2632be5ccecc44ef3851a649128ebc5346e884772cfca8155bb4604a68df97"},"schema_version":"1.0","source":{"id":"2201.10838","kind":"arxiv","version":18}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2201.10838","created_at":"2026-06-02T02:04:44Z"},{"alias_kind":"arxiv_version","alias_value":"2201.10838v18","created_at":"2026-06-02T02:04:44Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2201.10838","created_at":"2026-06-02T02:04:44Z"},{"alias_kind":"pith_short_12","alias_value":"ZICGYLA2UWRQ","created_at":"2026-06-02T02:04:44Z"},{"alias_kind":"pith_short_16","alias_value":"ZICGYLA2UWRQ2YHZ","created_at":"2026-06-02T02:04:44Z"},{"alias_kind":"pith_short_8","alias_value":"ZICGYLA2","created_at":"2026-06-02T02:04:44Z"}],"graph_snapshots":[{"event_id":"sha256:cf009398a7a091aaee3123d1ab06a3748d39a9ddf4babb8a33e8a4216913dc55","target":"graph","created_at":"2026-06-02T02:04:44Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":4,"items":[{"attestation":"unclaimed","claim_id":"C1","kind":"strongest_claim","source":"verdict.strongest_claim","status":"machine_extracted","text":"By incorporating this quadratic gradient, we enhance Nesterov's Accelerated Gradient (NAG), Adaptive Gradient (AdaGrad), and Adam algorithms... achieving comparable performance within only four iterations."},{"attestation":"unclaimed","claim_id":"C2","kind":"weakest_assumption","source":"verdict.weakest_assumption","status":"machine_extracted","text":"The quadratic gradient can be computed and applied under homomorphic encryption with negligible extra cost while preserving the claimed convergence properties and numerical stability across datasets."},{"attestation":"unclaimed","claim_id":"C3","kind":"one_line_summary","source":"verdict.one_line_summary","status":"machine_extracted","text":"Introduces quadratic gradient to enhance NAG, AdaGrad and Adam for logistic regression, claiming SOTA convergence and comparable encrypted training performance in four iterations."},{"attestation":"unclaimed","claim_id":"C4","kind":"headline","source":"verdict.pith_extraction.headline","status":"machine_extracted","text":"A quadratic gradient variant lets enhanced optimizers train logistic regression on encrypted data with plaintext-level performance in four iterations."}],"snapshot_sha256":"abde91407acae049091d0ad112541aa8a09c07fbfb1de12a884433f0b7bf8748"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2201.10838/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Training logistic regression over encrypted data has emerged as a prominent approach to addressing security concerns in recent years. In this paper, we introduce an efficient gradient variant, termed the \\textit{quadratic gradient}, which is specifically designed for privacy-preserving logistic regression while remaining equally effective in plaintext optimization. By incorporating this quadratic gradient, we enhance Nesterov's Accelerated Gradient (NAG), Adaptive Gradient (AdaGrad), and Adam algorithms. We evaluate these enhanced algorithms across various datasets, with experimental results d","authors_text":"John Chiang","cross_cats":["cs.LG"],"headline":"A quadratic gradient variant lets enhanced optimizers train logistic regression on encrypted data with plaintext-level performance in four iterations.","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CR","submitted_at":"2022-01-26T09:44:13Z","title":"Privacy-Preserving Logistic Regression Training with A Faster Gradient Variant"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2201.10838","kind":"arxiv","version":18},"verdict":{"created_at":"2026-05-24T12:17:18.779365Z","id":"63e81eef-8edc-4944-866a-444f50bf528d","model_set":{"reader":"grok-4.3"},"one_line_summary":"Introduces quadratic gradient to enhance NAG, AdaGrad and Adam for logistic regression, claiming SOTA convergence and comparable encrypted training performance in four iterations.","pipeline_version":"pith-pipeline@v0.9.0","pith_extraction_headline":"A quadratic gradient variant lets enhanced optimizers train logistic regression on encrypted data with plaintext-level performance in four iterations.","strongest_claim":"By incorporating this quadratic gradient, we enhance Nesterov's Accelerated Gradient (NAG), Adaptive Gradient (AdaGrad), and Adam algorithms... achieving comparable performance within only four iterations.","weakest_assumption":"The quadratic gradient can be computed and applied under homomorphic encryption with negligible extra cost while preserving the claimed convergence properties and numerical stability across datasets."}},"verdict_id":"63e81eef-8edc-4944-866a-444f50bf528d"}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:085a7ab50c9b3349b678355485d65a286d6f861086a142cd202da0759d7e8414","target":"record","created_at":"2026-06-02T02:04:44Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"8e05f06c6ff4bffd189f844c3ded8f1caf9282e5e402b767cbfed8894024ac6a","cross_cats_sorted":["cs.LG"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CR","submitted_at":"2022-01-26T09:44:13Z","title_canon_sha256":"1d2632be5ccecc44ef3851a649128ebc5346e884772cfca8155bb4604a68df97"},"schema_version":"1.0","source":{"id":"2201.10838","kind":"arxiv","version":18}},"canonical_sha256":"ca046c2c1aa5a30d60f9d0ba9f98a51008139ec40fc855d3305c5c63af310086","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"ca046c2c1aa5a30d60f9d0ba9f98a51008139ec40fc855d3305c5c63af310086","first_computed_at":"2026-06-02T02:04:44.486758Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-06-02T02:04:44.486758Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"LCHY+rBUAosWsgepj/EP+/ZTOdOkkbMKtnifFoUSZq78mfvLz3o90zFLzMrOwOO28LEVqZ40WRImbrbKTs3IAA==","signature_status":"signed_v1","signed_at":"2026-06-02T02:04:44.487147Z","signed_message":"canonical_sha256_bytes"},"source_id":"2201.10838","source_kind":"arxiv","source_version":18}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:085a7ab50c9b3349b678355485d65a286d6f861086a142cd202da0759d7e8414","sha256:cf009398a7a091aaee3123d1ab06a3748d39a9ddf4babb8a33e8a4216913dc55"],"state_sha256":"88dd2944ab2da4f6b3c377f73fe69d1201ee50abd01b7067f285b513a70450cb"}