{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2019:ZK7IU255YIAYPI4VEYVJF56DUC","short_pith_number":"pith:ZK7IU255","schema_version":"1.0","canonical_sha256":"cabe8a6bbdc20187a395262a92f7c3a0a70b3bdddff852615d077911228b0163","source":{"kind":"arxiv","id":"1909.05830","version":2},"attestation_state":"computed","paper":{"title":"Differentially Private Meta-Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.CR","stat.ML"],"primary_cat":"cs.LG","authors_text":"Ameet Talwalkar, Jeffrey Li, Mikhail Khodak, Sebastian Caldas","submitted_at":"2019-09-12T17:37:08Z","abstract_excerpt":"Parameter-transfer is a well-known and versatile approach for meta-learning, with applications including few-shot learning, federated learning, and reinforcement learning. However, parameter-transfer algorithms often require sharing models that have been trained on the samples from specific tasks, thus leaving the task-owners susceptible to breaches of privacy. We conduct the first formal study of privacy in this setting and formalize the notion of task-global differential privacy as a practical relaxation of more commonly studied threat models. We then propose a new differentially private alg"},"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":"1909.05830","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2019-09-12T17:37:08Z","cross_cats_sorted":["cs.AI","cs.CR","stat.ML"],"title_canon_sha256":"43d20558017283fca3a51cdda4230c8956fdd4d11c8015940d08799e3498651a","abstract_canon_sha256":"064b5e818ca0494341b8d4ae8e0ed4aaa9054cde750d8bd04d4de6b557f1160f"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T00:42:52.902510Z","signature_b64":"+C0u9TOloeixvS6WpDwKQLcmxvkw5GsJi3R5HWiEpE2QPHFfItQ4cf6kZ0rseLBrJMJbVgHJcr0I+qHuifrJDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"cabe8a6bbdc20187a395262a92f7c3a0a70b3bdddff852615d077911228b0163","last_reissued_at":"2026-07-05T00:42:52.902038Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T00:42:52.902038Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Differentially Private Meta-Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.CR","stat.ML"],"primary_cat":"cs.LG","authors_text":"Ameet Talwalkar, Jeffrey Li, Mikhail Khodak, Sebastian Caldas","submitted_at":"2019-09-12T17:37:08Z","abstract_excerpt":"Parameter-transfer is a well-known and versatile approach for meta-learning, with applications including few-shot learning, federated learning, and reinforcement learning. However, parameter-transfer algorithms often require sharing models that have been trained on the samples from specific tasks, thus leaving the task-owners susceptible to breaches of privacy. We conduct the first formal study of privacy in this setting and formalize the notion of task-global differential privacy as a practical relaxation of more commonly studied threat models. We then propose a new differentially private alg"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1909.05830","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/1909.05830/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":"1909.05830","created_at":"2026-07-05T00:42:52.902092+00:00"},{"alias_kind":"arxiv_version","alias_value":"1909.05830v2","created_at":"2026-07-05T00:42:52.902092+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1909.05830","created_at":"2026-07-05T00:42:52.902092+00:00"},{"alias_kind":"pith_short_12","alias_value":"ZK7IU255YIAY","created_at":"2026-07-05T00:42:52.902092+00:00"},{"alias_kind":"pith_short_16","alias_value":"ZK7IU255YIAYPI4V","created_at":"2026-07-05T00:42:52.902092+00:00"},{"alias_kind":"pith_short_8","alias_value":"ZK7IU255","created_at":"2026-07-05T00:42:52.902092+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2604.23426","citing_title":"Enhanced Privacy and Communication Efficiency in Non-IID Federated Learning with Adaptive Quantization and Differential Privacy","ref_index":27,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/ZK7IU255YIAYPI4VEYVJF56DUC","json":"https://pith.science/pith/ZK7IU255YIAYPI4VEYVJF56DUC.json","graph_json":"https://pith.science/api/pith-number/ZK7IU255YIAYPI4VEYVJF56DUC/graph.json","events_json":"https://pith.science/api/pith-number/ZK7IU255YIAYPI4VEYVJF56DUC/events.json","paper":"https://pith.science/paper/ZK7IU255"},"agent_actions":{"view_html":"https://pith.science/pith/ZK7IU255YIAYPI4VEYVJF56DUC","download_json":"https://pith.science/pith/ZK7IU255YIAYPI4VEYVJF56DUC.json","view_paper":"https://pith.science/paper/ZK7IU255","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1909.05830&json=true","fetch_graph":"https://pith.science/api/pith-number/ZK7IU255YIAYPI4VEYVJF56DUC/graph.json","fetch_events":"https://pith.science/api/pith-number/ZK7IU255YIAYPI4VEYVJF56DUC/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/ZK7IU255YIAYPI4VEYVJF56DUC/action/timestamp_anchor","attest_storage":"https://pith.science/pith/ZK7IU255YIAYPI4VEYVJF56DUC/action/storage_attestation","attest_author":"https://pith.science/pith/ZK7IU255YIAYPI4VEYVJF56DUC/action/author_attestation","sign_citation":"https://pith.science/pith/ZK7IU255YIAYPI4VEYVJF56DUC/action/citation_signature","submit_replication":"https://pith.science/pith/ZK7IU255YIAYPI4VEYVJF56DUC/action/replication_record"}},"created_at":"2026-07-05T00:42:52.902092+00:00","updated_at":"2026-07-05T00:42:52.902092+00:00"}