{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:C4QAOTHPDVBS3UO62BBFD2OOFH","short_pith_number":"pith:C4QAOTHP","schema_version":"1.0","canonical_sha256":"1720074cef1d432dd1ded04251e9ce29dee2db6a1b1f3bff91efaebe00e47111","source":{"kind":"arxiv","id":"2201.02265","version":1},"attestation_state":"computed","paper":{"title":"Learning to be adversarially robust and differentially private","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Borja Balle, Jamie Hayes, M. Pawan Kumar","submitted_at":"2022-01-06T22:33:06Z","abstract_excerpt":"We study the difficulties in learning that arise from robust and differentially private optimization. We first study convergence of gradient descent based adversarial training with differential privacy, taking a simple binary classification task on linearly separable data as an illustrative example. We compare the gap between adversarial and nominal risk in both private and non-private settings, showing that the data dimensionality dependent term introduced by private optimization compounds the difficulties of learning a robust model. After this, we discuss what parts of adversarial training a"},"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":"2201.02265","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2022-01-06T22:33:06Z","cross_cats_sorted":[],"title_canon_sha256":"2cb839db4a40aabe23ef061cfb4b485911ae931046bc33b1c099717a6c2198fd","abstract_canon_sha256":"e207a54cbd78482edc5cb7222ff724d82823a6cdbc55727c8f3dbe810d30846c"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T03:46:39.442078Z","signature_b64":"BCS1BmP6aoU0siiVCNTtHBlGR0UnegcHA4HedTo5jOET4tzblrEb0/JfSwrFt6yMLx6k9X4eSDn3tN/q7uZ5Bw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"1720074cef1d432dd1ded04251e9ce29dee2db6a1b1f3bff91efaebe00e47111","last_reissued_at":"2026-07-05T03:46:39.441653Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T03:46:39.441653Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Learning to be adversarially robust and differentially private","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Borja Balle, Jamie Hayes, M. Pawan Kumar","submitted_at":"2022-01-06T22:33:06Z","abstract_excerpt":"We study the difficulties in learning that arise from robust and differentially private optimization. We first study convergence of gradient descent based adversarial training with differential privacy, taking a simple binary classification task on linearly separable data as an illustrative example. We compare the gap between adversarial and nominal risk in both private and non-private settings, showing that the data dimensionality dependent term introduced by private optimization compounds the difficulties of learning a robust model. After this, we discuss what parts of adversarial training a"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2201.02265","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/2201.02265/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":"2201.02265","created_at":"2026-07-05T03:46:39.441714+00:00"},{"alias_kind":"arxiv_version","alias_value":"2201.02265v1","created_at":"2026-07-05T03:46:39.441714+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2201.02265","created_at":"2026-07-05T03:46:39.441714+00:00"},{"alias_kind":"pith_short_12","alias_value":"C4QAOTHPDVBS","created_at":"2026-07-05T03:46:39.441714+00:00"},{"alias_kind":"pith_short_16","alias_value":"C4QAOTHPDVBS3UO6","created_at":"2026-07-05T03:46:39.441714+00:00"},{"alias_kind":"pith_short_8","alias_value":"C4QAOTHP","created_at":"2026-07-05T03:46:39.441714+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.27148","citing_title":"Landseer: Exploring the Machine Learning Defense Landscape","ref_index":43,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/C4QAOTHPDVBS3UO62BBFD2OOFH","json":"https://pith.science/pith/C4QAOTHPDVBS3UO62BBFD2OOFH.json","graph_json":"https://pith.science/api/pith-number/C4QAOTHPDVBS3UO62BBFD2OOFH/graph.json","events_json":"https://pith.science/api/pith-number/C4QAOTHPDVBS3UO62BBFD2OOFH/events.json","paper":"https://pith.science/paper/C4QAOTHP"},"agent_actions":{"view_html":"https://pith.science/pith/C4QAOTHPDVBS3UO62BBFD2OOFH","download_json":"https://pith.science/pith/C4QAOTHPDVBS3UO62BBFD2OOFH.json","view_paper":"https://pith.science/paper/C4QAOTHP","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2201.02265&json=true","fetch_graph":"https://pith.science/api/pith-number/C4QAOTHPDVBS3UO62BBFD2OOFH/graph.json","fetch_events":"https://pith.science/api/pith-number/C4QAOTHPDVBS3UO62BBFD2OOFH/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/C4QAOTHPDVBS3UO62BBFD2OOFH/action/timestamp_anchor","attest_storage":"https://pith.science/pith/C4QAOTHPDVBS3UO62BBFD2OOFH/action/storage_attestation","attest_author":"https://pith.science/pith/C4QAOTHPDVBS3UO62BBFD2OOFH/action/author_attestation","sign_citation":"https://pith.science/pith/C4QAOTHPDVBS3UO62BBFD2OOFH/action/citation_signature","submit_replication":"https://pith.science/pith/C4QAOTHPDVBS3UO62BBFD2OOFH/action/replication_record"}},"created_at":"2026-07-05T03:46:39.441714+00:00","updated_at":"2026-07-05T03:46:39.441714+00:00"}