{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:V55E4CNRMONP3ARMZTZLF2BBRZ","short_pith_number":"pith:V55E4CNR","schema_version":"1.0","canonical_sha256":"af7a4e09b1639afd822cccf2b2e8218e78118920fb696d99e102238cad27b4e1","source":{"kind":"arxiv","id":"2305.16173","version":3},"attestation_state":"computed","paper":{"title":"Efficient Bound of Lipschitz Constant for Convolutional Layers by Gram Iteration","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Alexandre Allauzen, Alexandre Araujo, Blaise Delattre, Quentin Barth\\'elemy","submitted_at":"2023-05-25T15:32:21Z","abstract_excerpt":"Since the control of the Lipschitz constant has a great impact on the training stability, generalization, and robustness of neural networks, the estimation of this value is nowadays a real scientific challenge. In this paper we introduce a precise, fast, and differentiable upper bound for the spectral norm of convolutional layers using circulant matrix theory and a new alternative to the Power iteration. Called the Gram iteration, our approach exhibits a superlinear convergence. First, we show through a comprehensive set of experiments that our approach outperforms other state-of-the-art metho"},"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":"2305.16173","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2023-05-25T15:32:21Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"5851d69e35bd12ed05c3d8405cba9bf71a43c16c110e6b99b3f63bb8e99eed86","abstract_canon_sha256":"130784251fc6832294015c92b875d4700e23ba34bd70951e37d4ab1c81b9e094"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T06:22:05.657313Z","signature_b64":"vKceoG7+U1FgpbdNrkEb/jBO2M1JfS+QaCfceUS3bXvCDzIJLwzAAMLrejMWI6JXlswPD4odPK9tJYL5xVdBDw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"af7a4e09b1639afd822cccf2b2e8218e78118920fb696d99e102238cad27b4e1","last_reissued_at":"2026-07-05T06:22:05.656941Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T06:22:05.656941Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Efficient Bound of Lipschitz Constant for Convolutional Layers by Gram Iteration","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Alexandre Allauzen, Alexandre Araujo, Blaise Delattre, Quentin Barth\\'elemy","submitted_at":"2023-05-25T15:32:21Z","abstract_excerpt":"Since the control of the Lipschitz constant has a great impact on the training stability, generalization, and robustness of neural networks, the estimation of this value is nowadays a real scientific challenge. In this paper we introduce a precise, fast, and differentiable upper bound for the spectral norm of convolutional layers using circulant matrix theory and a new alternative to the Power iteration. Called the Gram iteration, our approach exhibits a superlinear convergence. First, we show through a comprehensive set of experiments that our approach outperforms other state-of-the-art metho"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2305.16173","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/2305.16173/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":"2305.16173","created_at":"2026-07-05T06:22:05.657006+00:00"},{"alias_kind":"arxiv_version","alias_value":"2305.16173v3","created_at":"2026-07-05T06:22:05.657006+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2305.16173","created_at":"2026-07-05T06:22:05.657006+00:00"},{"alias_kind":"pith_short_12","alias_value":"V55E4CNRMONP","created_at":"2026-07-05T06:22:05.657006+00:00"},{"alias_kind":"pith_short_16","alias_value":"V55E4CNRMONP3ARM","created_at":"2026-07-05T06:22:05.657006+00:00"},{"alias_kind":"pith_short_8","alias_value":"V55E4CNR","created_at":"2026-07-05T06:22:05.657006+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/V55E4CNRMONP3ARMZTZLF2BBRZ","json":"https://pith.science/pith/V55E4CNRMONP3ARMZTZLF2BBRZ.json","graph_json":"https://pith.science/api/pith-number/V55E4CNRMONP3ARMZTZLF2BBRZ/graph.json","events_json":"https://pith.science/api/pith-number/V55E4CNRMONP3ARMZTZLF2BBRZ/events.json","paper":"https://pith.science/paper/V55E4CNR"},"agent_actions":{"view_html":"https://pith.science/pith/V55E4CNRMONP3ARMZTZLF2BBRZ","download_json":"https://pith.science/pith/V55E4CNRMONP3ARMZTZLF2BBRZ.json","view_paper":"https://pith.science/paper/V55E4CNR","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2305.16173&json=true","fetch_graph":"https://pith.science/api/pith-number/V55E4CNRMONP3ARMZTZLF2BBRZ/graph.json","fetch_events":"https://pith.science/api/pith-number/V55E4CNRMONP3ARMZTZLF2BBRZ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/V55E4CNRMONP3ARMZTZLF2BBRZ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/V55E4CNRMONP3ARMZTZLF2BBRZ/action/storage_attestation","attest_author":"https://pith.science/pith/V55E4CNRMONP3ARMZTZLF2BBRZ/action/author_attestation","sign_citation":"https://pith.science/pith/V55E4CNRMONP3ARMZTZLF2BBRZ/action/citation_signature","submit_replication":"https://pith.science/pith/V55E4CNRMONP3ARMZTZLF2BBRZ/action/replication_record"}},"created_at":"2026-07-05T06:22:05.657006+00:00","updated_at":"2026-07-05T06:22:05.657006+00:00"}