{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:QDE7ASTYIKID3XO2J2UH243N67","short_pith_number":"pith:QDE7ASTY","schema_version":"1.0","canonical_sha256":"80c9f04a7842903dddda4ea87d736df7ecb3ef050221ea8be837c7f60a3864fd","source":{"kind":"arxiv","id":"2103.11109","version":6},"attestation_state":"computed","paper":{"title":"DataLens: Scalable Privacy Preserving Training via Gradient Compression and Aggregation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CR"],"primary_cat":"cs.LG","authors_text":"Bo Li, Boxin Wang, Ce Zhang, Fan Wu, Luka Rimanic, Yunhui Long","submitted_at":"2021-03-20T06:14:19Z","abstract_excerpt":"Recent success of deep neural networks (DNNs) hinges on the availability of large-scale dataset; however, training on such dataset often poses privacy risks for sensitive training information. In this paper, we aim to explore the power of generative models and gradient sparsity, and propose a scalable privacy-preserving generative model DATALENS. Comparing with the standard PATE privacy-preserving framework which allows teachers to vote on one-dimensional predictions, voting on the high dimensional gradient vectors is challenging in terms of privacy preservation. As dimension reduction techniq"},"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":"2103.11109","kind":"arxiv","version":6},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2021-03-20T06:14:19Z","cross_cats_sorted":["cs.CR"],"title_canon_sha256":"fc0c801ffcf22b69e3866f0c02a5b22711a3b81f8f42d09b5f8e03562d661c68","abstract_canon_sha256":"69729d85cbf50adb78a52b55f1f85446ac441e1c190b36f45356d71d82f4ab29"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T04:08:56.756639Z","signature_b64":"I4+6vuriYAEGGw5YMyq5voti3CX0X6/jhgWzB1j1Xyy1GMtkfkty0qsMEYWungjRES7zTO14MqBKmm+4LkYmAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"80c9f04a7842903dddda4ea87d736df7ecb3ef050221ea8be837c7f60a3864fd","last_reissued_at":"2026-07-05T04:08:56.756183Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T04:08:56.756183Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"DataLens: Scalable Privacy Preserving Training via Gradient Compression and Aggregation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CR"],"primary_cat":"cs.LG","authors_text":"Bo Li, Boxin Wang, Ce Zhang, Fan Wu, Luka Rimanic, Yunhui Long","submitted_at":"2021-03-20T06:14:19Z","abstract_excerpt":"Recent success of deep neural networks (DNNs) hinges on the availability of large-scale dataset; however, training on such dataset often poses privacy risks for sensitive training information. In this paper, we aim to explore the power of generative models and gradient sparsity, and propose a scalable privacy-preserving generative model DATALENS. Comparing with the standard PATE privacy-preserving framework which allows teachers to vote on one-dimensional predictions, voting on the high dimensional gradient vectors is challenging in terms of privacy preservation. As dimension reduction techniq"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2103.11109","kind":"arxiv","version":6},"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/2103.11109/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":"2103.11109","created_at":"2026-07-05T04:08:56.756240+00:00"},{"alias_kind":"arxiv_version","alias_value":"2103.11109v6","created_at":"2026-07-05T04:08:56.756240+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2103.11109","created_at":"2026-07-05T04:08:56.756240+00:00"},{"alias_kind":"pith_short_12","alias_value":"QDE7ASTYIKID","created_at":"2026-07-05T04:08:56.756240+00:00"},{"alias_kind":"pith_short_16","alias_value":"QDE7ASTYIKID3XO2","created_at":"2026-07-05T04:08:56.756240+00:00"},{"alias_kind":"pith_short_8","alias_value":"QDE7ASTY","created_at":"2026-07-05T04:08:56.756240+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/QDE7ASTYIKID3XO2J2UH243N67","json":"https://pith.science/pith/QDE7ASTYIKID3XO2J2UH243N67.json","graph_json":"https://pith.science/api/pith-number/QDE7ASTYIKID3XO2J2UH243N67/graph.json","events_json":"https://pith.science/api/pith-number/QDE7ASTYIKID3XO2J2UH243N67/events.json","paper":"https://pith.science/paper/QDE7ASTY"},"agent_actions":{"view_html":"https://pith.science/pith/QDE7ASTYIKID3XO2J2UH243N67","download_json":"https://pith.science/pith/QDE7ASTYIKID3XO2J2UH243N67.json","view_paper":"https://pith.science/paper/QDE7ASTY","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2103.11109&json=true","fetch_graph":"https://pith.science/api/pith-number/QDE7ASTYIKID3XO2J2UH243N67/graph.json","fetch_events":"https://pith.science/api/pith-number/QDE7ASTYIKID3XO2J2UH243N67/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/QDE7ASTYIKID3XO2J2UH243N67/action/timestamp_anchor","attest_storage":"https://pith.science/pith/QDE7ASTYIKID3XO2J2UH243N67/action/storage_attestation","attest_author":"https://pith.science/pith/QDE7ASTYIKID3XO2J2UH243N67/action/author_attestation","sign_citation":"https://pith.science/pith/QDE7ASTYIKID3XO2J2UH243N67/action/citation_signature","submit_replication":"https://pith.science/pith/QDE7ASTYIKID3XO2J2UH243N67/action/replication_record"}},"created_at":"2026-07-05T04:08:56.756240+00:00","updated_at":"2026-07-05T04:08:56.756240+00:00"}