{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2017:3OHGEFNWCZKVXR365AGCM37TI3","short_pith_number":"pith:3OHGEFNW","schema_version":"1.0","canonical_sha256":"db8e6215b616555bc77ee80c266ff346fca8d22968c94e97d12f0875392450d5","source":{"kind":"arxiv","id":"1711.00851","version":3},"attestation_state":"computed","paper":{"title":"Provable defenses against adversarial examples via the convex outer adversarial polytope","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","math.OC"],"primary_cat":"cs.LG","authors_text":"Eric Wong, J. Zico Kolter","submitted_at":"2017-11-02T17:59:24Z","abstract_excerpt":"We propose a method to learn deep ReLU-based classifiers that are provably robust against norm-bounded adversarial perturbations on the training data. For previously unseen examples, the approach is guaranteed to detect all adversarial examples, though it may flag some non-adversarial examples as well. The basic idea is to consider a convex outer approximation of the set of activations reachable through a norm-bounded perturbation, and we develop a robust optimization procedure that minimizes the worst case loss over this outer region (via a linear program). Crucially, we show that the dual pr"},"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":"1711.00851","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2017-11-02T17:59:24Z","cross_cats_sorted":["cs.AI","math.OC"],"title_canon_sha256":"811502c50d47f25ffa83c5757caa881e0abc21981460692b912146deafa0506b","abstract_canon_sha256":"0d9aa5f7c7893d5348d09a4d4e34b1af0d1a22a939e14a2680a33ddc4ae16712"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-18T00:13:47.382794Z","signature_b64":"y3gzDSAw0/e291SIARzJyDjbYDCw5UabR8P10aFVJt8FB7wSljvcR4PA3OzHXGV/NuaQY8A27j25hxX9aOjZDw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"db8e6215b616555bc77ee80c266ff346fca8d22968c94e97d12f0875392450d5","last_reissued_at":"2026-05-18T00:13:47.382167Z","signature_status":"signed_v1","first_computed_at":"2026-05-18T00:13:47.382167Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Provable defenses against adversarial examples via the convex outer adversarial polytope","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","math.OC"],"primary_cat":"cs.LG","authors_text":"Eric Wong, J. Zico Kolter","submitted_at":"2017-11-02T17:59:24Z","abstract_excerpt":"We propose a method to learn deep ReLU-based classifiers that are provably robust against norm-bounded adversarial perturbations on the training data. For previously unseen examples, the approach is guaranteed to detect all adversarial examples, though it may flag some non-adversarial examples as well. The basic idea is to consider a convex outer approximation of the set of activations reachable through a norm-bounded perturbation, and we develop a robust optimization procedure that minimizes the worst case loss over this outer region (via a linear program). Crucially, we show that the dual pr"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1711.00851","kind":"arxiv","version":3},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"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":"1711.00851","created_at":"2026-05-18T00:13:47.382255+00:00"},{"alias_kind":"arxiv_version","alias_value":"1711.00851v3","created_at":"2026-05-18T00:13:47.382255+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1711.00851","created_at":"2026-05-18T00:13:47.382255+00:00"},{"alias_kind":"pith_short_12","alias_value":"3OHGEFNWCZKV","created_at":"2026-05-18T12:30:58.224056+00:00"},{"alias_kind":"pith_short_16","alias_value":"3OHGEFNWCZKVXR36","created_at":"2026-05-18T12:30:58.224056+00:00"},{"alias_kind":"pith_short_8","alias_value":"3OHGEFNW","created_at":"2026-05-18T12:30:58.224056+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/3OHGEFNWCZKVXR365AGCM37TI3","json":"https://pith.science/pith/3OHGEFNWCZKVXR365AGCM37TI3.json","graph_json":"https://pith.science/api/pith-number/3OHGEFNWCZKVXR365AGCM37TI3/graph.json","events_json":"https://pith.science/api/pith-number/3OHGEFNWCZKVXR365AGCM37TI3/events.json","paper":"https://pith.science/paper/3OHGEFNW"},"agent_actions":{"view_html":"https://pith.science/pith/3OHGEFNWCZKVXR365AGCM37TI3","download_json":"https://pith.science/pith/3OHGEFNWCZKVXR365AGCM37TI3.json","view_paper":"https://pith.science/paper/3OHGEFNW","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1711.00851&json=true","fetch_graph":"https://pith.science/api/pith-number/3OHGEFNWCZKVXR365AGCM37TI3/graph.json","fetch_events":"https://pith.science/api/pith-number/3OHGEFNWCZKVXR365AGCM37TI3/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/3OHGEFNWCZKVXR365AGCM37TI3/action/timestamp_anchor","attest_storage":"https://pith.science/pith/3OHGEFNWCZKVXR365AGCM37TI3/action/storage_attestation","attest_author":"https://pith.science/pith/3OHGEFNWCZKVXR365AGCM37TI3/action/author_attestation","sign_citation":"https://pith.science/pith/3OHGEFNWCZKVXR365AGCM37TI3/action/citation_signature","submit_replication":"https://pith.science/pith/3OHGEFNWCZKVXR365AGCM37TI3/action/replication_record"}},"created_at":"2026-05-18T00:13:47.382255+00:00","updated_at":"2026-05-18T00:13:47.382255+00:00"}