{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:DBA3OPGFUOSMKSXOO5M3EJ2OND","short_pith_number":"pith:DBA3OPGF","schema_version":"1.0","canonical_sha256":"1841b73cc5a3a4c54aee7759b2274e68d4b99bb306443535ab7b958330b8ed25","source":{"kind":"arxiv","id":"2410.05733","version":2},"attestation_state":"computed","paper":{"title":"Private and Communication-Efficient Federated Learning based on Differentially Private Sketches","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CR"],"primary_cat":"cs.LG","authors_text":"Lihua Yin, Meifan Zhang, Zhanhong Xie","submitted_at":"2024-10-08T06:50:41Z","abstract_excerpt":"Federated learning (FL) faces two primary challenges: the risk of privacy leakage due to parameter sharing and communication inefficiencies. To address these challenges, we propose DPSFL, a federated learning method that utilizes differentially private sketches. DPSFL compresses the local gradients of each client using a count sketch, thereby improving communication efficiency, while adding noise to the sketches to ensure differential privacy (DP). We provide a theoretical analysis of privacy and convergence for the proposed method. Gradient clipping is essential in DP learning to limit sensit"},"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":"2410.05733","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2024-10-08T06:50:41Z","cross_cats_sorted":["cs.CR"],"title_canon_sha256":"43f1e0a9b073da189aa0f7ff0b2968b9434b1f79263bd544de0826ddba26eb54","abstract_canon_sha256":"915e8b2110a79b29f68460dc72914f8daf1c2aec1b741e7562bf60fbacea2e5d"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:18:31.427019Z","signature_b64":"w7smtzhtv9tMCos70thm3aREgrDsco4VU9iYkS1W5IVqXZulq6yV+8YGKTTxQaAph1CtVtgZ5e8hNa8EEp4XCg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"1841b73cc5a3a4c54aee7759b2274e68d4b99bb306443535ab7b958330b8ed25","last_reissued_at":"2026-07-05T09:18:31.426562Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:18:31.426562Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Private and Communication-Efficient Federated Learning based on Differentially Private Sketches","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CR"],"primary_cat":"cs.LG","authors_text":"Lihua Yin, Meifan Zhang, Zhanhong Xie","submitted_at":"2024-10-08T06:50:41Z","abstract_excerpt":"Federated learning (FL) faces two primary challenges: the risk of privacy leakage due to parameter sharing and communication inefficiencies. To address these challenges, we propose DPSFL, a federated learning method that utilizes differentially private sketches. DPSFL compresses the local gradients of each client using a count sketch, thereby improving communication efficiency, while adding noise to the sketches to ensure differential privacy (DP). We provide a theoretical analysis of privacy and convergence for the proposed method. Gradient clipping is essential in DP learning to limit sensit"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2410.05733","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/2410.05733/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":"2410.05733","created_at":"2026-07-05T09:18:31.426614+00:00"},{"alias_kind":"arxiv_version","alias_value":"2410.05733v2","created_at":"2026-07-05T09:18:31.426614+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2410.05733","created_at":"2026-07-05T09:18:31.426614+00:00"},{"alias_kind":"pith_short_12","alias_value":"DBA3OPGFUOSM","created_at":"2026-07-05T09:18:31.426614+00:00"},{"alias_kind":"pith_short_16","alias_value":"DBA3OPGFUOSMKSXO","created_at":"2026-07-05T09:18:31.426614+00:00"},{"alias_kind":"pith_short_8","alias_value":"DBA3OPGF","created_at":"2026-07-05T09:18:31.426614+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2509.08195","citing_title":"Sketched Gaussian Mechanism for Private Federated Learning","ref_index":2,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/DBA3OPGFUOSMKSXOO5M3EJ2OND","json":"https://pith.science/pith/DBA3OPGFUOSMKSXOO5M3EJ2OND.json","graph_json":"https://pith.science/api/pith-number/DBA3OPGFUOSMKSXOO5M3EJ2OND/graph.json","events_json":"https://pith.science/api/pith-number/DBA3OPGFUOSMKSXOO5M3EJ2OND/events.json","paper":"https://pith.science/paper/DBA3OPGF"},"agent_actions":{"view_html":"https://pith.science/pith/DBA3OPGFUOSMKSXOO5M3EJ2OND","download_json":"https://pith.science/pith/DBA3OPGFUOSMKSXOO5M3EJ2OND.json","view_paper":"https://pith.science/paper/DBA3OPGF","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2410.05733&json=true","fetch_graph":"https://pith.science/api/pith-number/DBA3OPGFUOSMKSXOO5M3EJ2OND/graph.json","fetch_events":"https://pith.science/api/pith-number/DBA3OPGFUOSMKSXOO5M3EJ2OND/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/DBA3OPGFUOSMKSXOO5M3EJ2OND/action/timestamp_anchor","attest_storage":"https://pith.science/pith/DBA3OPGFUOSMKSXOO5M3EJ2OND/action/storage_attestation","attest_author":"https://pith.science/pith/DBA3OPGFUOSMKSXOO5M3EJ2OND/action/author_attestation","sign_citation":"https://pith.science/pith/DBA3OPGFUOSMKSXOO5M3EJ2OND/action/citation_signature","submit_replication":"https://pith.science/pith/DBA3OPGFUOSMKSXOO5M3EJ2OND/action/replication_record"}},"created_at":"2026-07-05T09:18:31.426614+00:00","updated_at":"2026-07-05T09:18:31.426614+00:00"}