{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:4W7NEQZMYLUPUT4574QZ2LO45T","short_pith_number":"pith:4W7NEQZM","schema_version":"1.0","canonical_sha256":"e5bed2432cc2e8fa4f9dff219d2ddceccf86b1e124242e5e6f738145fa56edd4","source":{"kind":"arxiv","id":"2203.03106","version":3},"attestation_state":"computed","paper":{"title":"Differentially Private Federated Learning with Local Regularization and Sparsification","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CR","cs.CV"],"primary_cat":"cs.LG","authors_text":"Anda Cheng, Jian Cheng, Peisong Wang, Xi Sheryl Zhang","submitted_at":"2022-03-07T02:48:16Z","abstract_excerpt":"User-level differential privacy (DP) provides certifiable privacy guarantees to the information that is specific to any user's data in federated learning. Existing methods that ensure user-level DP come at the cost of severe accuracy decrease. In this paper, we study the cause of model performance degradation in federated learning under user-level DP guarantee. We find the key to solving this issue is to naturally restrict the norm of local updates before executing operations that guarantee DP. To this end, we propose two techniques, Bounded Local Update Regularization and Local Update Sparsif"},"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":"2203.03106","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2022-03-07T02:48:16Z","cross_cats_sorted":["cs.CR","cs.CV"],"title_canon_sha256":"9bc983ece0083fa93dbc664a0f93fdecd039011784d6b9c335109741f2d1716f","abstract_canon_sha256":"57cfc7345097411ba73438ea766dbb2dca6012897043d4edb06e8a7b3ea42c65"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T04:06:49.948956Z","signature_b64":"kIJEeLgrklhTPcOcI9CiurNslQm56g9YoEM4tho78LyKZhp2ybiNVx600fVtggRaWi0GgdyMb+61iWrjyew0CA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"e5bed2432cc2e8fa4f9dff219d2ddceccf86b1e124242e5e6f738145fa56edd4","last_reissued_at":"2026-07-05T04:06:49.948504Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T04:06:49.948504Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Differentially Private Federated Learning with Local Regularization and Sparsification","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CR","cs.CV"],"primary_cat":"cs.LG","authors_text":"Anda Cheng, Jian Cheng, Peisong Wang, Xi Sheryl Zhang","submitted_at":"2022-03-07T02:48:16Z","abstract_excerpt":"User-level differential privacy (DP) provides certifiable privacy guarantees to the information that is specific to any user's data in federated learning. Existing methods that ensure user-level DP come at the cost of severe accuracy decrease. In this paper, we study the cause of model performance degradation in federated learning under user-level DP guarantee. We find the key to solving this issue is to naturally restrict the norm of local updates before executing operations that guarantee DP. To this end, we propose two techniques, Bounded Local Update Regularization and Local Update Sparsif"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2203.03106","kind":"arxiv","version":3},"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/2203.03106/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":"2203.03106","created_at":"2026-07-05T04:06:49.948564+00:00"},{"alias_kind":"arxiv_version","alias_value":"2203.03106v3","created_at":"2026-07-05T04:06:49.948564+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2203.03106","created_at":"2026-07-05T04:06:49.948564+00:00"},{"alias_kind":"pith_short_12","alias_value":"4W7NEQZMYLUP","created_at":"2026-07-05T04:06:49.948564+00:00"},{"alias_kind":"pith_short_16","alias_value":"4W7NEQZMYLUPUT45","created_at":"2026-07-05T04:06:49.948564+00:00"},{"alias_kind":"pith_short_8","alias_value":"4W7NEQZM","created_at":"2026-07-05T04:06:49.948564+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2501.11659","citing_title":"BlindFL: Segmented Federated Learning with Fully Homomorphic Encryption","ref_index":5,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/4W7NEQZMYLUPUT4574QZ2LO45T","json":"https://pith.science/pith/4W7NEQZMYLUPUT4574QZ2LO45T.json","graph_json":"https://pith.science/api/pith-number/4W7NEQZMYLUPUT4574QZ2LO45T/graph.json","events_json":"https://pith.science/api/pith-number/4W7NEQZMYLUPUT4574QZ2LO45T/events.json","paper":"https://pith.science/paper/4W7NEQZM"},"agent_actions":{"view_html":"https://pith.science/pith/4W7NEQZMYLUPUT4574QZ2LO45T","download_json":"https://pith.science/pith/4W7NEQZMYLUPUT4574QZ2LO45T.json","view_paper":"https://pith.science/paper/4W7NEQZM","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2203.03106&json=true","fetch_graph":"https://pith.science/api/pith-number/4W7NEQZMYLUPUT4574QZ2LO45T/graph.json","fetch_events":"https://pith.science/api/pith-number/4W7NEQZMYLUPUT4574QZ2LO45T/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/4W7NEQZMYLUPUT4574QZ2LO45T/action/timestamp_anchor","attest_storage":"https://pith.science/pith/4W7NEQZMYLUPUT4574QZ2LO45T/action/storage_attestation","attest_author":"https://pith.science/pith/4W7NEQZMYLUPUT4574QZ2LO45T/action/author_attestation","sign_citation":"https://pith.science/pith/4W7NEQZMYLUPUT4574QZ2LO45T/action/citation_signature","submit_replication":"https://pith.science/pith/4W7NEQZMYLUPUT4574QZ2LO45T/action/replication_record"}},"created_at":"2026-07-05T04:06:49.948564+00:00","updated_at":"2026-07-05T04:06:49.948564+00:00"}