{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2020:VSKE26LQ2OKZGYFRSYVEUVMY77","short_pith_number":"pith:VSKE26LQ","schema_version":"1.0","canonical_sha256":"ac944d7970d3959360b1962a4a5598ffe2e0355f7d597be8d6cd74f335bcdd49","source":{"kind":"arxiv","id":"2006.06520","version":3},"attestation_state":"computed","paper":{"title":"Achieving robustness in classification using optimal transport with hinge regularization","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Alberto Gonz\\'alez-Sanz, Eustasio del Barrio, Franck Mamalet, Jean-Michel Loubes, Mathieu Serrurier, Thibaut Boissin","submitted_at":"2020-06-11T15:36:23Z","abstract_excerpt":"Adversarial examples have pointed out Deep Neural Networks vulnerability to small local noise. It has been shown that constraining their Lipschitz constant should enhance robustness, but make them harder to learn with classical loss functions. We propose a new framework for binary classification, based on optimal transport, which integrates this Lipschitz constraint as a theoretical requirement. We propose to learn 1-Lipschitz networks using a new loss that is an hinge regularized version of the Kantorovich-Rubinstein dual formulation for the Wasserstein distance estimation. This loss function"},"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":"2006.06520","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2020-06-11T15:36:23Z","cross_cats_sorted":["stat.ML"],"title_canon_sha256":"8c42a5f9a29bbe91301833f4506419fd9c3e3964e27132da6c44e8fbe3d2a45d","abstract_canon_sha256":"6f9e27af7ae543024cf4fb510b371ead0ea1cf2584f1d694e1368f5ab4c89ac4"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T02:34:40.113458Z","signature_b64":"rfefTTGo/q26T5l4xP/LmRBNnq4QfXzft4e6UagmKU61v4YEa6qkro5c3n69fvhp3pyafZPfKfLZKyVpxIlTDg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"ac944d7970d3959360b1962a4a5598ffe2e0355f7d597be8d6cd74f335bcdd49","last_reissued_at":"2026-07-05T02:34:40.112938Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T02:34:40.112938Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Achieving robustness in classification using optimal transport with hinge regularization","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Alberto Gonz\\'alez-Sanz, Eustasio del Barrio, Franck Mamalet, Jean-Michel Loubes, Mathieu Serrurier, Thibaut Boissin","submitted_at":"2020-06-11T15:36:23Z","abstract_excerpt":"Adversarial examples have pointed out Deep Neural Networks vulnerability to small local noise. It has been shown that constraining their Lipschitz constant should enhance robustness, but make them harder to learn with classical loss functions. We propose a new framework for binary classification, based on optimal transport, which integrates this Lipschitz constraint as a theoretical requirement. We propose to learn 1-Lipschitz networks using a new loss that is an hinge regularized version of the Kantorovich-Rubinstein dual formulation for the Wasserstein distance estimation. This loss function"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2006.06520","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":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2006.06520/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":"2006.06520","created_at":"2026-07-05T02:34:40.112994+00:00"},{"alias_kind":"arxiv_version","alias_value":"2006.06520v3","created_at":"2026-07-05T02:34:40.112994+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2006.06520","created_at":"2026-07-05T02:34:40.112994+00:00"},{"alias_kind":"pith_short_12","alias_value":"VSKE26LQ2OKZ","created_at":"2026-07-05T02:34:40.112994+00:00"},{"alias_kind":"pith_short_16","alias_value":"VSKE26LQ2OKZGYFR","created_at":"2026-07-05T02:34:40.112994+00:00"},{"alias_kind":"pith_short_8","alias_value":"VSKE26LQ","created_at":"2026-07-05T02:34:40.112994+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/VSKE26LQ2OKZGYFRSYVEUVMY77","json":"https://pith.science/pith/VSKE26LQ2OKZGYFRSYVEUVMY77.json","graph_json":"https://pith.science/api/pith-number/VSKE26LQ2OKZGYFRSYVEUVMY77/graph.json","events_json":"https://pith.science/api/pith-number/VSKE26LQ2OKZGYFRSYVEUVMY77/events.json","paper":"https://pith.science/paper/VSKE26LQ"},"agent_actions":{"view_html":"https://pith.science/pith/VSKE26LQ2OKZGYFRSYVEUVMY77","download_json":"https://pith.science/pith/VSKE26LQ2OKZGYFRSYVEUVMY77.json","view_paper":"https://pith.science/paper/VSKE26LQ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2006.06520&json=true","fetch_graph":"https://pith.science/api/pith-number/VSKE26LQ2OKZGYFRSYVEUVMY77/graph.json","fetch_events":"https://pith.science/api/pith-number/VSKE26LQ2OKZGYFRSYVEUVMY77/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/VSKE26LQ2OKZGYFRSYVEUVMY77/action/timestamp_anchor","attest_storage":"https://pith.science/pith/VSKE26LQ2OKZGYFRSYVEUVMY77/action/storage_attestation","attest_author":"https://pith.science/pith/VSKE26LQ2OKZGYFRSYVEUVMY77/action/author_attestation","sign_citation":"https://pith.science/pith/VSKE26LQ2OKZGYFRSYVEUVMY77/action/citation_signature","submit_replication":"https://pith.science/pith/VSKE26LQ2OKZGYFRSYVEUVMY77/action/replication_record"}},"created_at":"2026-07-05T02:34:40.112994+00:00","updated_at":"2026-07-05T02:34:40.112994+00:00"}