{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2018:3F2PZ3NJTUDYGUZPCBDTSU5V4Y","short_pith_number":"pith:3F2PZ3NJ","schema_version":"1.0","canonical_sha256":"d974fceda99d0783532f10473953b5e600c1b8342831fd4da6048467d7322828","source":{"kind":"arxiv","id":"1810.01075","version":1},"attestation_state":"computed","paper":{"title":"Implicit Self-Regularization in Deep Neural Networks: Evidence from Random Matrix Theory and Implications for Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Charles H. Martin, Michael W. Mahoney","submitted_at":"2018-10-02T05:27:59Z","abstract_excerpt":"Random Matrix Theory (RMT) is applied to analyze weight matrices of Deep Neural Networks (DNNs), including both production quality, pre-trained models such as AlexNet and Inception, and smaller models trained from scratch, such as LeNet5 and a miniature-AlexNet. Empirical and theoretical results clearly indicate that the DNN training process itself implicitly implements a form of Self-Regularization. The empirical spectral density (ESD) of DNN layer matrices displays signatures of traditionally-regularized statistical models, even in the absence of exogenously specifying traditional forms of e"},"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":"1810.01075","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2018-10-02T05:27:59Z","cross_cats_sorted":["stat.ML"],"title_canon_sha256":"0d9ae95a7b073ec29bd22ff9980e0689837a3f579ad1cfab97e6643c25469f59","abstract_canon_sha256":"9b2f3fd67add00664f04d66ff7b7020c1bca33da1e88ad21c66bc44d82ca8b38"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-18T00:04:18.933358Z","signature_b64":"GH2uPN5kUVJEsE+5u5EAVHjvfO+FPkO9hPNSYDdGBtaZ+SX5irIOQCkZbd4fCwhV3R7RmjX6VRc7Zu2XVEniAg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"d974fceda99d0783532f10473953b5e600c1b8342831fd4da6048467d7322828","last_reissued_at":"2026-05-18T00:04:18.932841Z","signature_status":"signed_v1","first_computed_at":"2026-05-18T00:04:18.932841Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Implicit Self-Regularization in Deep Neural Networks: Evidence from Random Matrix Theory and Implications for Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Charles H. Martin, Michael W. Mahoney","submitted_at":"2018-10-02T05:27:59Z","abstract_excerpt":"Random Matrix Theory (RMT) is applied to analyze weight matrices of Deep Neural Networks (DNNs), including both production quality, pre-trained models such as AlexNet and Inception, and smaller models trained from scratch, such as LeNet5 and a miniature-AlexNet. Empirical and theoretical results clearly indicate that the DNN training process itself implicitly implements a form of Self-Regularization. The empirical spectral density (ESD) of DNN layer matrices displays signatures of traditionally-regularized statistical models, even in the absence of exogenously specifying traditional forms of e"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1810.01075","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":"1810.01075","created_at":"2026-05-18T00:04:18.932913+00:00"},{"alias_kind":"arxiv_version","alias_value":"1810.01075v1","created_at":"2026-05-18T00:04:18.932913+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1810.01075","created_at":"2026-05-18T00:04:18.932913+00:00"},{"alias_kind":"pith_short_12","alias_value":"3F2PZ3NJTUDY","created_at":"2026-05-18T12:32:02.567920+00:00"},{"alias_kind":"pith_short_16","alias_value":"3F2PZ3NJTUDYGUZP","created_at":"2026-05-18T12:32:02.567920+00:00"},{"alias_kind":"pith_short_8","alias_value":"3F2PZ3NJ","created_at":"2026-05-18T12:32:02.567920+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2511.21715","citing_title":"DNNs, Dataset Statistics, and Correlation Functions","ref_index":19,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/3F2PZ3NJTUDYGUZPCBDTSU5V4Y","json":"https://pith.science/pith/3F2PZ3NJTUDYGUZPCBDTSU5V4Y.json","graph_json":"https://pith.science/api/pith-number/3F2PZ3NJTUDYGUZPCBDTSU5V4Y/graph.json","events_json":"https://pith.science/api/pith-number/3F2PZ3NJTUDYGUZPCBDTSU5V4Y/events.json","paper":"https://pith.science/paper/3F2PZ3NJ"},"agent_actions":{"view_html":"https://pith.science/pith/3F2PZ3NJTUDYGUZPCBDTSU5V4Y","download_json":"https://pith.science/pith/3F2PZ3NJTUDYGUZPCBDTSU5V4Y.json","view_paper":"https://pith.science/paper/3F2PZ3NJ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1810.01075&json=true","fetch_graph":"https://pith.science/api/pith-number/3F2PZ3NJTUDYGUZPCBDTSU5V4Y/graph.json","fetch_events":"https://pith.science/api/pith-number/3F2PZ3NJTUDYGUZPCBDTSU5V4Y/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/3F2PZ3NJTUDYGUZPCBDTSU5V4Y/action/timestamp_anchor","attest_storage":"https://pith.science/pith/3F2PZ3NJTUDYGUZPCBDTSU5V4Y/action/storage_attestation","attest_author":"https://pith.science/pith/3F2PZ3NJTUDYGUZPCBDTSU5V4Y/action/author_attestation","sign_citation":"https://pith.science/pith/3F2PZ3NJTUDYGUZPCBDTSU5V4Y/action/citation_signature","submit_replication":"https://pith.science/pith/3F2PZ3NJTUDYGUZPCBDTSU5V4Y/action/replication_record"}},"created_at":"2026-05-18T00:04:18.932913+00:00","updated_at":"2026-05-18T00:04:18.932913+00:00"}