{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2020:JOS2BVMXGDD5Y5WAHZ2GCBU5HC","short_pith_number":"pith:JOS2BVMX","schema_version":"1.0","canonical_sha256":"4ba5a0d59730c7dc76c03e7461069d38a66c985c6456453604477d0cc46ef740","source":{"kind":"arxiv","id":"2006.11601","version":2},"attestation_state":"computed","paper":{"title":"Rethinking Privacy Preserving Deep Learning: How to Evaluate and Thwart Privacy Attacks","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CR","cs.DC","stat.ML"],"primary_cat":"cs.LG","authors_text":"Ce Ju, Chang Liu, Chee Seng Chan, Kam Woh Ng, Lixin Fan, Qiang Yang, Tianyu Zhang","submitted_at":"2020-06-20T15:48:57Z","abstract_excerpt":"This paper investigates capabilities of Privacy-Preserving Deep Learning (PPDL) mechanisms against various forms of privacy attacks. First, we propose to quantitatively measure the trade-off between model accuracy and privacy losses incurred by reconstruction, tracing and membership attacks. Second, we formulate reconstruction attacks as solving a noisy system of linear equations, and prove that attacks are guaranteed to be defeated if condition (2) is unfulfilled. Third, based on theoretical analysis, a novel Secret Polarization Network (SPN) is proposed to thwart privacy attacks, which pose "},"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":"2006.11601","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2020-06-20T15:48:57Z","cross_cats_sorted":["cs.CR","cs.DC","stat.ML"],"title_canon_sha256":"a96331472fb7e9a9da781712b1df26647cb187dbb3461368648416a6bc70708c","abstract_canon_sha256":"f5de8b0b43493b6d766b7ee8c511bd7729ac7fa5dc3e712b26803392b29cc2a5"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T01:13:12.119516Z","signature_b64":"+232yDUuIfuFbWxffrEgxLCGwO5V9UaIV2D5ltAi+WycFzipxgpHIUA5D0DWUI3XDCfLuacOs+rOXnPl6N8GBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"4ba5a0d59730c7dc76c03e7461069d38a66c985c6456453604477d0cc46ef740","last_reissued_at":"2026-07-05T01:13:12.119071Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T01:13:12.119071Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Rethinking Privacy Preserving Deep Learning: How to Evaluate and Thwart Privacy Attacks","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CR","cs.DC","stat.ML"],"primary_cat":"cs.LG","authors_text":"Ce Ju, Chang Liu, Chee Seng Chan, Kam Woh Ng, Lixin Fan, Qiang Yang, Tianyu Zhang","submitted_at":"2020-06-20T15:48:57Z","abstract_excerpt":"This paper investigates capabilities of Privacy-Preserving Deep Learning (PPDL) mechanisms against various forms of privacy attacks. First, we propose to quantitatively measure the trade-off between model accuracy and privacy losses incurred by reconstruction, tracing and membership attacks. Second, we formulate reconstruction attacks as solving a noisy system of linear equations, and prove that attacks are guaranteed to be defeated if condition (2) is unfulfilled. Third, based on theoretical analysis, a novel Secret Polarization Network (SPN) is proposed to thwart privacy attacks, which pose "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2006.11601","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/2006.11601/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":"2006.11601","created_at":"2026-07-05T01:13:12.119121+00:00"},{"alias_kind":"arxiv_version","alias_value":"2006.11601v2","created_at":"2026-07-05T01:13:12.119121+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2006.11601","created_at":"2026-07-05T01:13:12.119121+00:00"},{"alias_kind":"pith_short_12","alias_value":"JOS2BVMXGDD5","created_at":"2026-07-05T01:13:12.119121+00:00"},{"alias_kind":"pith_short_16","alias_value":"JOS2BVMXGDD5Y5WA","created_at":"2026-07-05T01:13:12.119121+00:00"},{"alias_kind":"pith_short_8","alias_value":"JOS2BVMX","created_at":"2026-07-05T01:13:12.119121+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/JOS2BVMXGDD5Y5WAHZ2GCBU5HC","json":"https://pith.science/pith/JOS2BVMXGDD5Y5WAHZ2GCBU5HC.json","graph_json":"https://pith.science/api/pith-number/JOS2BVMXGDD5Y5WAHZ2GCBU5HC/graph.json","events_json":"https://pith.science/api/pith-number/JOS2BVMXGDD5Y5WAHZ2GCBU5HC/events.json","paper":"https://pith.science/paper/JOS2BVMX"},"agent_actions":{"view_html":"https://pith.science/pith/JOS2BVMXGDD5Y5WAHZ2GCBU5HC","download_json":"https://pith.science/pith/JOS2BVMXGDD5Y5WAHZ2GCBU5HC.json","view_paper":"https://pith.science/paper/JOS2BVMX","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2006.11601&json=true","fetch_graph":"https://pith.science/api/pith-number/JOS2BVMXGDD5Y5WAHZ2GCBU5HC/graph.json","fetch_events":"https://pith.science/api/pith-number/JOS2BVMXGDD5Y5WAHZ2GCBU5HC/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/JOS2BVMXGDD5Y5WAHZ2GCBU5HC/action/timestamp_anchor","attest_storage":"https://pith.science/pith/JOS2BVMXGDD5Y5WAHZ2GCBU5HC/action/storage_attestation","attest_author":"https://pith.science/pith/JOS2BVMXGDD5Y5WAHZ2GCBU5HC/action/author_attestation","sign_citation":"https://pith.science/pith/JOS2BVMXGDD5Y5WAHZ2GCBU5HC/action/citation_signature","submit_replication":"https://pith.science/pith/JOS2BVMXGDD5Y5WAHZ2GCBU5HC/action/replication_record"}},"created_at":"2026-07-05T01:13:12.119121+00:00","updated_at":"2026-07-05T01:13:12.119121+00:00"}