{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2016:SHPZM62UMAOZGEBZQ37SGCVUWM","short_pith_number":"pith:SHPZM62U","schema_version":"1.0","canonical_sha256":"91df967b54601d93103986ff230ab4b32f8e22bc5bdd47606f902e8e79aee306","source":{"kind":"arxiv","id":"1611.02345","version":3},"attestation_state":"computed","paper":{"title":"Neural Taylor Approximations: Convergence and Exploration in Rectifier Networks","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.NE","stat.ML"],"primary_cat":"cs.LG","authors_text":"Brian McWilliams, David Balduzzi, Tony Butler-Yeoman","submitted_at":"2016-11-07T23:47:05Z","abstract_excerpt":"Modern convolutional networks, incorporating rectifiers and max-pooling, are neither smooth nor convex; standard guarantees therefore do not apply. Nevertheless, methods from convex optimization such as gradient descent and Adam are widely used as building blocks for deep learning algorithms. This paper provides the first convergence guarantee applicable to modern convnets, which furthermore matches a lower bound for convex nonsmooth functions. The key technical tool is the neural Taylor approximation -- a straightforward application of Taylor expansions to neural networks -- and the associate"},"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":"1611.02345","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2016-11-07T23:47:05Z","cross_cats_sorted":["cs.NE","stat.ML"],"title_canon_sha256":"1ae6844c759f341ea96043435239275859d6747f722f993a81ef4bc7a66de33c","abstract_canon_sha256":"9f0d1ecde1987a3e932b4c0bfd6b1f9cac44391d47469a95485cb8c1761a9592"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-18T00:14:06.970027Z","signature_b64":"3mXFcxlnIh8FsiEWrOfx/60kf45YTHtv/TuiHxcsnu+Qq+VNGlQ5Z+WPPc67XESDLXlCsrBzV0f6gBZfLro/AQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"91df967b54601d93103986ff230ab4b32f8e22bc5bdd47606f902e8e79aee306","last_reissued_at":"2026-05-18T00:14:06.969564Z","signature_status":"signed_v1","first_computed_at":"2026-05-18T00:14:06.969564Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Neural Taylor Approximations: Convergence and Exploration in Rectifier Networks","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.NE","stat.ML"],"primary_cat":"cs.LG","authors_text":"Brian McWilliams, David Balduzzi, Tony Butler-Yeoman","submitted_at":"2016-11-07T23:47:05Z","abstract_excerpt":"Modern convolutional networks, incorporating rectifiers and max-pooling, are neither smooth nor convex; standard guarantees therefore do not apply. Nevertheless, methods from convex optimization such as gradient descent and Adam are widely used as building blocks for deep learning algorithms. This paper provides the first convergence guarantee applicable to modern convnets, which furthermore matches a lower bound for convex nonsmooth functions. The key technical tool is the neural Taylor approximation -- a straightforward application of Taylor expansions to neural networks -- and the associate"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1611.02345","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":"1611.02345","created_at":"2026-05-18T00:14:06.969639+00:00"},{"alias_kind":"arxiv_version","alias_value":"1611.02345v3","created_at":"2026-05-18T00:14:06.969639+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1611.02345","created_at":"2026-05-18T00:14:06.969639+00:00"},{"alias_kind":"pith_short_12","alias_value":"SHPZM62UMAOZ","created_at":"2026-05-18T12:30:44.179134+00:00"},{"alias_kind":"pith_short_16","alias_value":"SHPZM62UMAOZGEBZ","created_at":"2026-05-18T12:30:44.179134+00:00"},{"alias_kind":"pith_short_8","alias_value":"SHPZM62U","created_at":"2026-05-18T12:30:44.179134+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/SHPZM62UMAOZGEBZQ37SGCVUWM","json":"https://pith.science/pith/SHPZM62UMAOZGEBZQ37SGCVUWM.json","graph_json":"https://pith.science/api/pith-number/SHPZM62UMAOZGEBZQ37SGCVUWM/graph.json","events_json":"https://pith.science/api/pith-number/SHPZM62UMAOZGEBZQ37SGCVUWM/events.json","paper":"https://pith.science/paper/SHPZM62U"},"agent_actions":{"view_html":"https://pith.science/pith/SHPZM62UMAOZGEBZQ37SGCVUWM","download_json":"https://pith.science/pith/SHPZM62UMAOZGEBZQ37SGCVUWM.json","view_paper":"https://pith.science/paper/SHPZM62U","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1611.02345&json=true","fetch_graph":"https://pith.science/api/pith-number/SHPZM62UMAOZGEBZQ37SGCVUWM/graph.json","fetch_events":"https://pith.science/api/pith-number/SHPZM62UMAOZGEBZQ37SGCVUWM/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/SHPZM62UMAOZGEBZQ37SGCVUWM/action/timestamp_anchor","attest_storage":"https://pith.science/pith/SHPZM62UMAOZGEBZQ37SGCVUWM/action/storage_attestation","attest_author":"https://pith.science/pith/SHPZM62UMAOZGEBZQ37SGCVUWM/action/author_attestation","sign_citation":"https://pith.science/pith/SHPZM62UMAOZGEBZQ37SGCVUWM/action/citation_signature","submit_replication":"https://pith.science/pith/SHPZM62UMAOZGEBZQ37SGCVUWM/action/replication_record"}},"created_at":"2026-05-18T00:14:06.969639+00:00","updated_at":"2026-05-18T00:14:06.969639+00:00"}