{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2020:BSTI7K2HXXU2PCCGQMITWZ5B5Q","short_pith_number":"pith:BSTI7K2H","schema_version":"1.0","canonical_sha256":"0ca68fab47bde9a7884683113b67a1ec258314d1c0676712082437ad2518aa02","source":{"kind":"arxiv","id":"2002.08570","version":1},"attestation_state":"computed","paper":{"title":"Input Perturbation: A New Paradigm between Central and Local Differential Privacy","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Ben Niu, Likun Zhang, Weiping Wang, Xinyi Tong, Yilin Kang, Yong Liu","submitted_at":"2020-02-20T05:20:02Z","abstract_excerpt":"Traditionally, there are two models on differential privacy: the central model and the local model. The central model focuses on the machine learning model and the local model focuses on the training data. In this paper, we study the \\textit{input perturbation} method in differentially private empirical risk minimization (DP-ERM), preserving privacy of the central model. By adding noise to the original training data and training with the `perturbed data', we achieve ($\\epsilon$,$\\delta$)-differential privacy on the final model, along with some kind of privacy on the original data. We observe t"},"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":"2002.08570","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2020-02-20T05:20:02Z","cross_cats_sorted":["stat.ML"],"title_canon_sha256":"9c51bcf00f97a2f645a091587cd06c68f6df283d36652d16b08852b01107797c","abstract_canon_sha256":"1a86e2cdb475f2ca6b0b3b30dbb517448e249ee964b7403f16761e96b7d807e4"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T00:42:36.215976Z","signature_b64":"+7UyPmJyPDXhjsnc3tZ1uiuVyzSqGqksP6+JLw5cMqOHUqrodVwc3brp8m4zsh5tNoQKp2W8it+qGar8zy/JCg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"0ca68fab47bde9a7884683113b67a1ec258314d1c0676712082437ad2518aa02","last_reissued_at":"2026-07-05T00:42:36.215521Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T00:42:36.215521Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Input Perturbation: A New Paradigm between Central and Local Differential Privacy","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Ben Niu, Likun Zhang, Weiping Wang, Xinyi Tong, Yilin Kang, Yong Liu","submitted_at":"2020-02-20T05:20:02Z","abstract_excerpt":"Traditionally, there are two models on differential privacy: the central model and the local model. The central model focuses on the machine learning model and the local model focuses on the training data. In this paper, we study the \\textit{input perturbation} method in differentially private empirical risk minimization (DP-ERM), preserving privacy of the central model. By adding noise to the original training data and training with the `perturbed data', we achieve ($\\epsilon$,$\\delta$)-differential privacy on the final model, along with some kind of privacy on the original data. We observe t"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2002.08570","kind":"arxiv","version":1},"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/2002.08570/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":"2002.08570","created_at":"2026-07-05T00:42:36.215574+00:00"},{"alias_kind":"arxiv_version","alias_value":"2002.08570v1","created_at":"2026-07-05T00:42:36.215574+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2002.08570","created_at":"2026-07-05T00:42:36.215574+00:00"},{"alias_kind":"pith_short_12","alias_value":"BSTI7K2HXXU2","created_at":"2026-07-05T00:42:36.215574+00:00"},{"alias_kind":"pith_short_16","alias_value":"BSTI7K2HXXU2PCCG","created_at":"2026-07-05T00:42:36.215574+00:00"},{"alias_kind":"pith_short_8","alias_value":"BSTI7K2H","created_at":"2026-07-05T00:42:36.215574+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.03945","citing_title":"Integrating Feature Correlation in Differential Privacy with Applications in DP-ERM","ref_index":11,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/BSTI7K2HXXU2PCCGQMITWZ5B5Q","json":"https://pith.science/pith/BSTI7K2HXXU2PCCGQMITWZ5B5Q.json","graph_json":"https://pith.science/api/pith-number/BSTI7K2HXXU2PCCGQMITWZ5B5Q/graph.json","events_json":"https://pith.science/api/pith-number/BSTI7K2HXXU2PCCGQMITWZ5B5Q/events.json","paper":"https://pith.science/paper/BSTI7K2H"},"agent_actions":{"view_html":"https://pith.science/pith/BSTI7K2HXXU2PCCGQMITWZ5B5Q","download_json":"https://pith.science/pith/BSTI7K2HXXU2PCCGQMITWZ5B5Q.json","view_paper":"https://pith.science/paper/BSTI7K2H","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2002.08570&json=true","fetch_graph":"https://pith.science/api/pith-number/BSTI7K2HXXU2PCCGQMITWZ5B5Q/graph.json","fetch_events":"https://pith.science/api/pith-number/BSTI7K2HXXU2PCCGQMITWZ5B5Q/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/BSTI7K2HXXU2PCCGQMITWZ5B5Q/action/timestamp_anchor","attest_storage":"https://pith.science/pith/BSTI7K2HXXU2PCCGQMITWZ5B5Q/action/storage_attestation","attest_author":"https://pith.science/pith/BSTI7K2HXXU2PCCGQMITWZ5B5Q/action/author_attestation","sign_citation":"https://pith.science/pith/BSTI7K2HXXU2PCCGQMITWZ5B5Q/action/citation_signature","submit_replication":"https://pith.science/pith/BSTI7K2HXXU2PCCGQMITWZ5B5Q/action/replication_record"}},"created_at":"2026-07-05T00:42:36.215574+00:00","updated_at":"2026-07-05T00:42:36.215574+00:00"}