{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:2GIIE3TMCHTFBRDJ26PR2XEPPN","short_pith_number":"pith:2GIIE3TM","schema_version":"1.0","canonical_sha256":"d190826e6c11e650c469d79f1d5c8f7b7079f5147462b8a90b6065808d4be60e","source":{"kind":"arxiv","id":"2204.06233","version":1},"attestation_state":"computed","paper":{"title":"Approximation of Lipschitz Functions using Deep Spline Neural Networks","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["math.OC"],"primary_cat":"cs.LG","authors_text":"Alexis Goujon, Michael Unser, Pakshal Bohra, Sebastian Neumayer","submitted_at":"2022-04-13T08:07:28Z","abstract_excerpt":"Lipschitz-constrained neural networks have many applications in machine learning. Since designing and training expressive Lipschitz-constrained networks is very challenging, there is a need for improved methods and a better theoretical understanding. Unfortunately, it turns out that ReLU networks have provable disadvantages in this setting. Hence, we propose to use learnable spline activation functions with at least 3 linear regions instead. We prove that this choice is optimal among all component-wise $1$-Lipschitz activation functions in the sense that no other weight constrained architectur"},"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":"2204.06233","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2022-04-13T08:07:28Z","cross_cats_sorted":["math.OC"],"title_canon_sha256":"d3399a3c1dce0cc7e65655fcf09c4576157270b947bd25a3c92575b60bec308b","abstract_canon_sha256":"e176131960e3507191a0df4c93e5bba3c054e268afa30b801de146592319f2f3"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T04:14:31.778064Z","signature_b64":"KqqXTKhWmFUFfRwSpCfJY9xvAgxQQ6lyCj2C7fDGLnc36NDw6GHNGYp8HDgL5uVQM/vU9J9uUeMzbOXQ2QT6Dg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"d190826e6c11e650c469d79f1d5c8f7b7079f5147462b8a90b6065808d4be60e","last_reissued_at":"2026-07-05T04:14:31.777567Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T04:14:31.777567Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Approximation of Lipschitz Functions using Deep Spline Neural Networks","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["math.OC"],"primary_cat":"cs.LG","authors_text":"Alexis Goujon, Michael Unser, Pakshal Bohra, Sebastian Neumayer","submitted_at":"2022-04-13T08:07:28Z","abstract_excerpt":"Lipschitz-constrained neural networks have many applications in machine learning. Since designing and training expressive Lipschitz-constrained networks is very challenging, there is a need for improved methods and a better theoretical understanding. Unfortunately, it turns out that ReLU networks have provable disadvantages in this setting. Hence, we propose to use learnable spline activation functions with at least 3 linear regions instead. We prove that this choice is optimal among all component-wise $1$-Lipschitz activation functions in the sense that no other weight constrained architectur"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2204.06233","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/2204.06233/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":"2204.06233","created_at":"2026-07-05T04:14:31.777632+00:00"},{"alias_kind":"arxiv_version","alias_value":"2204.06233v1","created_at":"2026-07-05T04:14:31.777632+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2204.06233","created_at":"2026-07-05T04:14:31.777632+00:00"},{"alias_kind":"pith_short_12","alias_value":"2GIIE3TMCHTF","created_at":"2026-07-05T04:14:31.777632+00:00"},{"alias_kind":"pith_short_16","alias_value":"2GIIE3TMCHTFBRDJ","created_at":"2026-07-05T04:14:31.777632+00:00"},{"alias_kind":"pith_short_8","alias_value":"2GIIE3TM","created_at":"2026-07-05T04:14:31.777632+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/2GIIE3TMCHTFBRDJ26PR2XEPPN","json":"https://pith.science/pith/2GIIE3TMCHTFBRDJ26PR2XEPPN.json","graph_json":"https://pith.science/api/pith-number/2GIIE3TMCHTFBRDJ26PR2XEPPN/graph.json","events_json":"https://pith.science/api/pith-number/2GIIE3TMCHTFBRDJ26PR2XEPPN/events.json","paper":"https://pith.science/paper/2GIIE3TM"},"agent_actions":{"view_html":"https://pith.science/pith/2GIIE3TMCHTFBRDJ26PR2XEPPN","download_json":"https://pith.science/pith/2GIIE3TMCHTFBRDJ26PR2XEPPN.json","view_paper":"https://pith.science/paper/2GIIE3TM","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2204.06233&json=true","fetch_graph":"https://pith.science/api/pith-number/2GIIE3TMCHTFBRDJ26PR2XEPPN/graph.json","fetch_events":"https://pith.science/api/pith-number/2GIIE3TMCHTFBRDJ26PR2XEPPN/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/2GIIE3TMCHTFBRDJ26PR2XEPPN/action/timestamp_anchor","attest_storage":"https://pith.science/pith/2GIIE3TMCHTFBRDJ26PR2XEPPN/action/storage_attestation","attest_author":"https://pith.science/pith/2GIIE3TMCHTFBRDJ26PR2XEPPN/action/author_attestation","sign_citation":"https://pith.science/pith/2GIIE3TMCHTFBRDJ26PR2XEPPN/action/citation_signature","submit_replication":"https://pith.science/pith/2GIIE3TMCHTFBRDJ26PR2XEPPN/action/replication_record"}},"created_at":"2026-07-05T04:14:31.777632+00:00","updated_at":"2026-07-05T04:14:31.777632+00:00"}