{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:ZIXNE4LKT7DJUWT3LVTZDKSQAC","short_pith_number":"pith:ZIXNE4LK","schema_version":"1.0","canonical_sha256":"ca2ed2716a9fc69a5a7b5d6791aa500087181b804bfd9ad57af3d9536a85639a","source":{"kind":"arxiv","id":"2512.13853","version":2},"attestation_state":"computed","paper":{"title":"Dropout Neural Network Training Viewed from a Percolation Perspective","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cond-mat.stat-mech","math.PR","stat.ML"],"primary_cat":"cs.LG","authors_text":"Finley Devlin, Jaron Sanders","submitted_at":"2025-12-15T19:39:25Z","abstract_excerpt":"In this work, we investigate the existence and effect of percolation in training deep Neural Networks (NNs) with dropout. Dropout methods are regularisation techniques for training NNs, first introduced by G. Hinton et al. (2012). These methods temporarily remove connections in the NN, randomly at each stage of training, and update the remaining subnetwork with Stochastic Gradient Descent (SGD). The process of removing connections from a network at random is similar to percolation, a paradigm model of statistical physics.\n  If dropout were to remove enough connections such that there is no pat"},"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":"2512.13853","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-12-15T19:39:25Z","cross_cats_sorted":["cond-mat.stat-mech","math.PR","stat.ML"],"title_canon_sha256":"df1725b2919ecb2a279202e1fc246001e32973f12b454e5c1386d0eff79e1f44","abstract_canon_sha256":"79a6579ef5c2027d9a710b11cc0dc4eff79d21b70100f8625b99abc69400d0c7"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-19T16:09:53.148250Z","signature_b64":"Rlbhr2aU7qxFttxaZgJYuS+FdVvQR7hRuaKTvTq2b173R9hemcyAFJRLUR+gOk6GCXrOcHWMdz6W0ix5hgT/DA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"ca2ed2716a9fc69a5a7b5d6791aa500087181b804bfd9ad57af3d9536a85639a","last_reissued_at":"2026-06-19T16:09:53.147780Z","signature_status":"signed_v1","first_computed_at":"2026-06-19T16:09:53.147780Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Dropout Neural Network Training Viewed from a Percolation Perspective","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cond-mat.stat-mech","math.PR","stat.ML"],"primary_cat":"cs.LG","authors_text":"Finley Devlin, Jaron Sanders","submitted_at":"2025-12-15T19:39:25Z","abstract_excerpt":"In this work, we investigate the existence and effect of percolation in training deep Neural Networks (NNs) with dropout. Dropout methods are regularisation techniques for training NNs, first introduced by G. Hinton et al. (2012). These methods temporarily remove connections in the NN, randomly at each stage of training, and update the remaining subnetwork with Stochastic Gradient Descent (SGD). The process of removing connections from a network at random is similar to percolation, a paradigm model of statistical physics.\n  If dropout were to remove enough connections such that there is no pat"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2512.13853","kind":"arxiv","version":2},"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/2512.13853/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":"2512.13853","created_at":"2026-06-19T16:09:53.147839+00:00"},{"alias_kind":"arxiv_version","alias_value":"2512.13853v2","created_at":"2026-06-19T16:09:53.147839+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2512.13853","created_at":"2026-06-19T16:09:53.147839+00:00"},{"alias_kind":"pith_short_12","alias_value":"ZIXNE4LKT7DJ","created_at":"2026-06-19T16:09:53.147839+00:00"},{"alias_kind":"pith_short_16","alias_value":"ZIXNE4LKT7DJUWT3","created_at":"2026-06-19T16:09:53.147839+00:00"},{"alias_kind":"pith_short_8","alias_value":"ZIXNE4LK","created_at":"2026-06-19T16:09:53.147839+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2606.20347","citing_title":"Critical Percolation as a Synthetic Data Model for Interpretability","ref_index":18,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/ZIXNE4LKT7DJUWT3LVTZDKSQAC","json":"https://pith.science/pith/ZIXNE4LKT7DJUWT3LVTZDKSQAC.json","graph_json":"https://pith.science/api/pith-number/ZIXNE4LKT7DJUWT3LVTZDKSQAC/graph.json","events_json":"https://pith.science/api/pith-number/ZIXNE4LKT7DJUWT3LVTZDKSQAC/events.json","paper":"https://pith.science/paper/ZIXNE4LK"},"agent_actions":{"view_html":"https://pith.science/pith/ZIXNE4LKT7DJUWT3LVTZDKSQAC","download_json":"https://pith.science/pith/ZIXNE4LKT7DJUWT3LVTZDKSQAC.json","view_paper":"https://pith.science/paper/ZIXNE4LK","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2512.13853&json=true","fetch_graph":"https://pith.science/api/pith-number/ZIXNE4LKT7DJUWT3LVTZDKSQAC/graph.json","fetch_events":"https://pith.science/api/pith-number/ZIXNE4LKT7DJUWT3LVTZDKSQAC/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/ZIXNE4LKT7DJUWT3LVTZDKSQAC/action/timestamp_anchor","attest_storage":"https://pith.science/pith/ZIXNE4LKT7DJUWT3LVTZDKSQAC/action/storage_attestation","attest_author":"https://pith.science/pith/ZIXNE4LKT7DJUWT3LVTZDKSQAC/action/author_attestation","sign_citation":"https://pith.science/pith/ZIXNE4LKT7DJUWT3LVTZDKSQAC/action/citation_signature","submit_replication":"https://pith.science/pith/ZIXNE4LKT7DJUWT3LVTZDKSQAC/action/replication_record"}},"created_at":"2026-06-19T16:09:53.147839+00:00","updated_at":"2026-06-19T16:09:53.147839+00:00"}