{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:7KWE4MEFPWCODOVR2X4RHWQJVU","short_pith_number":"pith:7KWE4MEF","schema_version":"1.0","canonical_sha256":"faac4e30857d84e1bab1d5f913da09ad08aa53b43524969bd555bd565cc3429d","source":{"kind":"arxiv","id":"2305.00873","version":2},"attestation_state":"computed","paper":{"title":"Towards the Flatter Landscape and Better Generalization in Federated Learning under Client-level Differential Privacy","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CR","cs.DC"],"primary_cat":"cs.LG","authors_text":"Bo Yuan, Dacheng Tao, Kang Wei, Li Shen, Xueqian Wang, Yifan Shi, Yingqi Liu","submitted_at":"2023-05-01T15:19:09Z","abstract_excerpt":"To defend the inference attacks and mitigate the sensitive information leakages in Federated Learning (FL), client-level Differentially Private FL (DPFL) is the de-facto standard for privacy protection by clipping local updates and adding random noise. However, existing DPFL methods tend to make a sharp loss landscape and have poor weight perturbation robustness, resulting in severe performance degradation. To alleviate these issues, we propose a novel DPFL algorithm named DP-FedSAM, which leverages gradient perturbation to mitigate the negative impact of DP. Specifically, DP-FedSAM integrates"},"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":"2305.00873","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2023-05-01T15:19:09Z","cross_cats_sorted":["cs.CR","cs.DC"],"title_canon_sha256":"a781c38b7bc9098e5ca312f66cece195e94cc5e7b86211fef9eb0b741dee281d","abstract_canon_sha256":"3d92ca4f66888b483dc51cc615891287ae1196854781e8cf27cab4e8b3421e78"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T06:06:20.006214Z","signature_b64":"WsS9mdD+f/W6Gtt8N3FxPTd7hAsSQFex2ozLJF60aeEQpMej8jDRLDDRmFzz5lIcWZ2sAdY7kbyy7v1ierK+Bg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"faac4e30857d84e1bab1d5f913da09ad08aa53b43524969bd555bd565cc3429d","last_reissued_at":"2026-07-05T06:06:20.005703Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T06:06:20.005703Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Towards the Flatter Landscape and Better Generalization in Federated Learning under Client-level Differential Privacy","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CR","cs.DC"],"primary_cat":"cs.LG","authors_text":"Bo Yuan, Dacheng Tao, Kang Wei, Li Shen, Xueqian Wang, Yifan Shi, Yingqi Liu","submitted_at":"2023-05-01T15:19:09Z","abstract_excerpt":"To defend the inference attacks and mitigate the sensitive information leakages in Federated Learning (FL), client-level Differentially Private FL (DPFL) is the de-facto standard for privacy protection by clipping local updates and adding random noise. However, existing DPFL methods tend to make a sharp loss landscape and have poor weight perturbation robustness, resulting in severe performance degradation. To alleviate these issues, we propose a novel DPFL algorithm named DP-FedSAM, which leverages gradient perturbation to mitigate the negative impact of DP. Specifically, DP-FedSAM integrates"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2305.00873","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/2305.00873/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":"2305.00873","created_at":"2026-07-05T06:06:20.005762+00:00"},{"alias_kind":"arxiv_version","alias_value":"2305.00873v2","created_at":"2026-07-05T06:06:20.005762+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2305.00873","created_at":"2026-07-05T06:06:20.005762+00:00"},{"alias_kind":"pith_short_12","alias_value":"7KWE4MEFPWCO","created_at":"2026-07-05T06:06:20.005762+00:00"},{"alias_kind":"pith_short_16","alias_value":"7KWE4MEFPWCODOVR","created_at":"2026-07-05T06:06:20.005762+00:00"},{"alias_kind":"pith_short_8","alias_value":"7KWE4MEF","created_at":"2026-07-05T06:06:20.005762+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2509.05377","citing_title":"Enhancing Gradient Variance and Differential Privacy in Quantum Federated Learning","ref_index":22,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/7KWE4MEFPWCODOVR2X4RHWQJVU","json":"https://pith.science/pith/7KWE4MEFPWCODOVR2X4RHWQJVU.json","graph_json":"https://pith.science/api/pith-number/7KWE4MEFPWCODOVR2X4RHWQJVU/graph.json","events_json":"https://pith.science/api/pith-number/7KWE4MEFPWCODOVR2X4RHWQJVU/events.json","paper":"https://pith.science/paper/7KWE4MEF"},"agent_actions":{"view_html":"https://pith.science/pith/7KWE4MEFPWCODOVR2X4RHWQJVU","download_json":"https://pith.science/pith/7KWE4MEFPWCODOVR2X4RHWQJVU.json","view_paper":"https://pith.science/paper/7KWE4MEF","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2305.00873&json=true","fetch_graph":"https://pith.science/api/pith-number/7KWE4MEFPWCODOVR2X4RHWQJVU/graph.json","fetch_events":"https://pith.science/api/pith-number/7KWE4MEFPWCODOVR2X4RHWQJVU/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/7KWE4MEFPWCODOVR2X4RHWQJVU/action/timestamp_anchor","attest_storage":"https://pith.science/pith/7KWE4MEFPWCODOVR2X4RHWQJVU/action/storage_attestation","attest_author":"https://pith.science/pith/7KWE4MEFPWCODOVR2X4RHWQJVU/action/author_attestation","sign_citation":"https://pith.science/pith/7KWE4MEFPWCODOVR2X4RHWQJVU/action/citation_signature","submit_replication":"https://pith.science/pith/7KWE4MEFPWCODOVR2X4RHWQJVU/action/replication_record"}},"created_at":"2026-07-05T06:06:20.005762+00:00","updated_at":"2026-07-05T06:06:20.005762+00:00"}