{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:SAPC7NGZJJLCLOWR34CU6CMJOE","short_pith_number":"pith:SAPC7NGZ","schema_version":"1.0","canonical_sha256":"901e2fb4d94a5625bad1df054f0989713cfdc30d4f2d448c8c245ace3a168449","source":{"kind":"arxiv","id":"2503.18317","version":1},"attestation_state":"computed","paper":{"title":"Improved Rates of Differentially Private Nonconvex-Strongly-Concave Minimax Optimization","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Di Wang, Jinhui Xu, Meng Ding, Mingxi Lei, Ruijia Zhang, Zihang Xiang","submitted_at":"2025-03-24T03:51:27Z","abstract_excerpt":"In this paper, we study the problem of (finite sum) minimax optimization in the Differential Privacy (DP) model. Unlike most of the previous studies on the (strongly) convex-concave settings or loss functions satisfying the Polyak-Lojasiewicz condition, here we mainly focus on the nonconvex-strongly-concave one, which encapsulates many models in deep learning such as deep AUC maximization. Specifically, we first analyze a DP version of Stochastic Gradient Descent Ascent (SGDA) and show that it is possible to get a DP estimator whose $l_2$-norm of the gradient for the empirical risk function is"},"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":"2503.18317","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-03-24T03:51:27Z","cross_cats_sorted":[],"title_canon_sha256":"3d7a4f7a2be90a1cf2a66559d66daf7e9cf13c4780a0f0e4253b16ab0b3fc3c9","abstract_canon_sha256":"57161e52035200a2fa44ce591d56d840d949e6a48fc40788f99cc81bf5835ec3"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:38:08.166433Z","signature_b64":"MQjx401H9FNkbG3ZaPIXkTqHhRQR6iE5kFtFCqfThAgcsiRlCuyAjKpXV2Na4BD0oMTANijLhk607f4exhfyAg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"901e2fb4d94a5625bad1df054f0989713cfdc30d4f2d448c8c245ace3a168449","last_reissued_at":"2026-07-05T10:38:08.165685Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:38:08.165685Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Improved Rates of Differentially Private Nonconvex-Strongly-Concave Minimax Optimization","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Di Wang, Jinhui Xu, Meng Ding, Mingxi Lei, Ruijia Zhang, Zihang Xiang","submitted_at":"2025-03-24T03:51:27Z","abstract_excerpt":"In this paper, we study the problem of (finite sum) minimax optimization in the Differential Privacy (DP) model. Unlike most of the previous studies on the (strongly) convex-concave settings or loss functions satisfying the Polyak-Lojasiewicz condition, here we mainly focus on the nonconvex-strongly-concave one, which encapsulates many models in deep learning such as deep AUC maximization. Specifically, we first analyze a DP version of Stochastic Gradient Descent Ascent (SGDA) and show that it is possible to get a DP estimator whose $l_2$-norm of the gradient for the empirical risk function is"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2503.18317","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/2503.18317/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":"2503.18317","created_at":"2026-07-05T10:38:08.165821+00:00"},{"alias_kind":"arxiv_version","alias_value":"2503.18317v1","created_at":"2026-07-05T10:38:08.165821+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2503.18317","created_at":"2026-07-05T10:38:08.165821+00:00"},{"alias_kind":"pith_short_12","alias_value":"SAPC7NGZJJLC","created_at":"2026-07-05T10:38:08.165821+00:00"},{"alias_kind":"pith_short_16","alias_value":"SAPC7NGZJJLCLOWR","created_at":"2026-07-05T10:38:08.165821+00:00"},{"alias_kind":"pith_short_8","alias_value":"SAPC7NGZ","created_at":"2026-07-05T10:38:08.165821+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/SAPC7NGZJJLCLOWR34CU6CMJOE","json":"https://pith.science/pith/SAPC7NGZJJLCLOWR34CU6CMJOE.json","graph_json":"https://pith.science/api/pith-number/SAPC7NGZJJLCLOWR34CU6CMJOE/graph.json","events_json":"https://pith.science/api/pith-number/SAPC7NGZJJLCLOWR34CU6CMJOE/events.json","paper":"https://pith.science/paper/SAPC7NGZ"},"agent_actions":{"view_html":"https://pith.science/pith/SAPC7NGZJJLCLOWR34CU6CMJOE","download_json":"https://pith.science/pith/SAPC7NGZJJLCLOWR34CU6CMJOE.json","view_paper":"https://pith.science/paper/SAPC7NGZ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2503.18317&json=true","fetch_graph":"https://pith.science/api/pith-number/SAPC7NGZJJLCLOWR34CU6CMJOE/graph.json","fetch_events":"https://pith.science/api/pith-number/SAPC7NGZJJLCLOWR34CU6CMJOE/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/SAPC7NGZJJLCLOWR34CU6CMJOE/action/timestamp_anchor","attest_storage":"https://pith.science/pith/SAPC7NGZJJLCLOWR34CU6CMJOE/action/storage_attestation","attest_author":"https://pith.science/pith/SAPC7NGZJJLCLOWR34CU6CMJOE/action/author_attestation","sign_citation":"https://pith.science/pith/SAPC7NGZJJLCLOWR34CU6CMJOE/action/citation_signature","submit_replication":"https://pith.science/pith/SAPC7NGZJJLCLOWR34CU6CMJOE/action/replication_record"}},"created_at":"2026-07-05T10:38:08.165821+00:00","updated_at":"2026-07-05T10:38:08.165821+00:00"}