{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2019:DTRYQU5OC3T3UHLTFDMYGUAVHI","short_pith_number":"pith:DTRYQU5O","schema_version":"1.0","canonical_sha256":"1ce38853ae16e7ba1d7328d98350153a39f020de90b0be62f1e0115be72ab04f","source":{"kind":"arxiv","id":"1902.10416","version":1},"attestation_state":"computed","paper":{"title":"Equi-normalization of Neural Networks","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CV","authors_text":"Benjamin Graham, Herv\\'e J\\'egou, Pierre Stock, R\\'emi Gribonval","submitted_at":"2019-02-27T09:52:43Z","abstract_excerpt":"Modern neural networks are over-parametrized. In particular, each rectified linear hidden unit can be modified by a multiplicative factor by adjusting input and output weights, without changing the rest of the network. Inspired by the Sinkhorn-Knopp algorithm, we introduce a fast iterative method for minimizing the L2 norm of the weights, equivalently the weight decay regularizer. It provably converges to a unique solution. Interleaving our algorithm with SGD during training improves the test accuracy. For small batches, our approach offers an alternative to batch-and group-normalization on CI"},"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":"1902.10416","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2019-02-27T09:52:43Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"34f01bb15a7a5aea80ad112991b5cf235c0a3befcb91c5ff66d114cb660d4051","abstract_canon_sha256":"b17ddbc6743fd1ce8a12a9333267d294835728f3d95dbf685ca3d2d1da517f21"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-17T23:52:30.898984Z","signature_b64":"zm2yd95meJGsi+C9b+p9VYwMj0VGU04MdW0JcrRMMYFHGMahNYMD5omOTljOpXXehbMLdUzOcp5Z8ZEN2ltgAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"1ce38853ae16e7ba1d7328d98350153a39f020de90b0be62f1e0115be72ab04f","last_reissued_at":"2026-05-17T23:52:30.898372Z","signature_status":"signed_v1","first_computed_at":"2026-05-17T23:52:30.898372Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Equi-normalization of Neural Networks","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CV","authors_text":"Benjamin Graham, Herv\\'e J\\'egou, Pierre Stock, R\\'emi Gribonval","submitted_at":"2019-02-27T09:52:43Z","abstract_excerpt":"Modern neural networks are over-parametrized. In particular, each rectified linear hidden unit can be modified by a multiplicative factor by adjusting input and output weights, without changing the rest of the network. Inspired by the Sinkhorn-Knopp algorithm, we introduce a fast iterative method for minimizing the L2 norm of the weights, equivalently the weight decay regularizer. It provably converges to a unique solution. Interleaving our algorithm with SGD during training improves the test accuracy. For small batches, our approach offers an alternative to batch-and group-normalization on CI"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1902.10416","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":""},"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":"1902.10416","created_at":"2026-05-17T23:52:30.898505+00:00"},{"alias_kind":"arxiv_version","alias_value":"1902.10416v1","created_at":"2026-05-17T23:52:30.898505+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1902.10416","created_at":"2026-05-17T23:52:30.898505+00:00"},{"alias_kind":"pith_short_12","alias_value":"DTRYQU5OC3T3","created_at":"2026-05-18T12:33:15.570797+00:00"},{"alias_kind":"pith_short_16","alias_value":"DTRYQU5OC3T3UHLT","created_at":"2026-05-18T12:33:15.570797+00:00"},{"alias_kind":"pith_short_8","alias_value":"DTRYQU5O","created_at":"2026-05-18T12:33:15.570797+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2604.22838","citing_title":"Neural Network Optimization Reimagined: Decoupled Techniques for Scratch and Fine-Tuning","ref_index":24,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/DTRYQU5OC3T3UHLTFDMYGUAVHI","json":"https://pith.science/pith/DTRYQU5OC3T3UHLTFDMYGUAVHI.json","graph_json":"https://pith.science/api/pith-number/DTRYQU5OC3T3UHLTFDMYGUAVHI/graph.json","events_json":"https://pith.science/api/pith-number/DTRYQU5OC3T3UHLTFDMYGUAVHI/events.json","paper":"https://pith.science/paper/DTRYQU5O"},"agent_actions":{"view_html":"https://pith.science/pith/DTRYQU5OC3T3UHLTFDMYGUAVHI","download_json":"https://pith.science/pith/DTRYQU5OC3T3UHLTFDMYGUAVHI.json","view_paper":"https://pith.science/paper/DTRYQU5O","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1902.10416&json=true","fetch_graph":"https://pith.science/api/pith-number/DTRYQU5OC3T3UHLTFDMYGUAVHI/graph.json","fetch_events":"https://pith.science/api/pith-number/DTRYQU5OC3T3UHLTFDMYGUAVHI/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/DTRYQU5OC3T3UHLTFDMYGUAVHI/action/timestamp_anchor","attest_storage":"https://pith.science/pith/DTRYQU5OC3T3UHLTFDMYGUAVHI/action/storage_attestation","attest_author":"https://pith.science/pith/DTRYQU5OC3T3UHLTFDMYGUAVHI/action/author_attestation","sign_citation":"https://pith.science/pith/DTRYQU5OC3T3UHLTFDMYGUAVHI/action/citation_signature","submit_replication":"https://pith.science/pith/DTRYQU5OC3T3UHLTFDMYGUAVHI/action/replication_record"}},"created_at":"2026-05-17T23:52:30.898505+00:00","updated_at":"2026-05-17T23:52:30.898505+00:00"}