{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:V5OYWKTCWBV76CAEQUMX3EG7WG","short_pith_number":"pith:V5OYWKTC","schema_version":"1.0","canonical_sha256":"af5d8b2a62b06bff080485197d90dfb18e2e4345509c2c3c83b78dfce28e4163","source":{"kind":"arxiv","id":"2108.12978","version":3},"attestation_state":"computed","paper":{"title":"Private Multi-Task Learning: Formulation and Applications to Federated Learning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CR"],"primary_cat":"cs.LG","authors_text":"Shengyuan Hu, Virginia Smith, Zhiwei Steven Wu","submitted_at":"2021-08-30T03:37:36Z","abstract_excerpt":"Many problems in machine learning rely on multi-task learning (MTL), in which the goal is to solve multiple related machine learning tasks simultaneously. MTL is particularly relevant for privacy-sensitive applications in areas such as healthcare, finance, and IoT computing, where sensitive data from multiple, varied sources are shared for the purpose of learning. In this work, we formalize notions of client-level privacy for MTL via joint differential privacy (JDP), a relaxation of differential privacy for mechanism design and distributed optimization. We then propose an algorithm for mean-re"},"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":"2108.12978","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2021-08-30T03:37:36Z","cross_cats_sorted":["cs.CR"],"title_canon_sha256":"c6428374d04f2b5750355a8ace6d8a1f0be60443d278281b36a78f034a090e38","abstract_canon_sha256":"fd62fbdd1b6ff3cd3b9e7e94c8083a4f0a46dcfb6fa308ff34e141e9e673f9b0"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:01:35.974618Z","signature_b64":"zyvVX9/tVjngnC7iFur76JeAzyzdBE+Tw9J7sDdeTtbFdlDgZaDmOe1jmZmeKklnn20ZLYOXIPNhb+ISB8UmDw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"af5d8b2a62b06bff080485197d90dfb18e2e4345509c2c3c83b78dfce28e4163","last_reissued_at":"2026-07-05T07:01:35.974182Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:01:35.974182Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Private Multi-Task Learning: Formulation and Applications to Federated Learning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CR"],"primary_cat":"cs.LG","authors_text":"Shengyuan Hu, Virginia Smith, Zhiwei Steven Wu","submitted_at":"2021-08-30T03:37:36Z","abstract_excerpt":"Many problems in machine learning rely on multi-task learning (MTL), in which the goal is to solve multiple related machine learning tasks simultaneously. MTL is particularly relevant for privacy-sensitive applications in areas such as healthcare, finance, and IoT computing, where sensitive data from multiple, varied sources are shared for the purpose of learning. In this work, we formalize notions of client-level privacy for MTL via joint differential privacy (JDP), a relaxation of differential privacy for mechanism design and distributed optimization. We then propose an algorithm for mean-re"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2108.12978","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/2108.12978/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":"2108.12978","created_at":"2026-07-05T07:01:35.974242+00:00"},{"alias_kind":"arxiv_version","alias_value":"2108.12978v3","created_at":"2026-07-05T07:01:35.974242+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2108.12978","created_at":"2026-07-05T07:01:35.974242+00:00"},{"alias_kind":"pith_short_12","alias_value":"V5OYWKTCWBV7","created_at":"2026-07-05T07:01:35.974242+00:00"},{"alias_kind":"pith_short_16","alias_value":"V5OYWKTCWBV76CAE","created_at":"2026-07-05T07:01:35.974242+00:00"},{"alias_kind":"pith_short_8","alias_value":"V5OYWKTC","created_at":"2026-07-05T07:01:35.974242+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2501.13904","citing_title":"Privacy-Preserving Personalized Federated Prompt Learning for Multimodal Large Language Models","ref_index":24,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/V5OYWKTCWBV76CAEQUMX3EG7WG","json":"https://pith.science/pith/V5OYWKTCWBV76CAEQUMX3EG7WG.json","graph_json":"https://pith.science/api/pith-number/V5OYWKTCWBV76CAEQUMX3EG7WG/graph.json","events_json":"https://pith.science/api/pith-number/V5OYWKTCWBV76CAEQUMX3EG7WG/events.json","paper":"https://pith.science/paper/V5OYWKTC"},"agent_actions":{"view_html":"https://pith.science/pith/V5OYWKTCWBV76CAEQUMX3EG7WG","download_json":"https://pith.science/pith/V5OYWKTCWBV76CAEQUMX3EG7WG.json","view_paper":"https://pith.science/paper/V5OYWKTC","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2108.12978&json=true","fetch_graph":"https://pith.science/api/pith-number/V5OYWKTCWBV76CAEQUMX3EG7WG/graph.json","fetch_events":"https://pith.science/api/pith-number/V5OYWKTCWBV76CAEQUMX3EG7WG/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/V5OYWKTCWBV76CAEQUMX3EG7WG/action/timestamp_anchor","attest_storage":"https://pith.science/pith/V5OYWKTCWBV76CAEQUMX3EG7WG/action/storage_attestation","attest_author":"https://pith.science/pith/V5OYWKTCWBV76CAEQUMX3EG7WG/action/author_attestation","sign_citation":"https://pith.science/pith/V5OYWKTCWBV76CAEQUMX3EG7WG/action/citation_signature","submit_replication":"https://pith.science/pith/V5OYWKTCWBV76CAEQUMX3EG7WG/action/replication_record"}},"created_at":"2026-07-05T07:01:35.974242+00:00","updated_at":"2026-07-05T07:01:35.974242+00:00"}