{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:VCTXVKTHUHPN42THYL2I4ZNMTK","short_pith_number":"pith:VCTXVKTH","schema_version":"1.0","canonical_sha256":"a8a77aaa67a1dede6a67c2f48e65ac9a83270d171075537bb9947ed699f8dbd5","source":{"kind":"arxiv","id":"2311.15365","version":2},"attestation_state":"computed","paper":{"title":"A convergence result of a continuous model of deep learning via \\L{}ojasiewicz--Simon inequality","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["math.AP","math.FA","math.PR"],"primary_cat":"cs.LG","authors_text":"Noboru Isobe","submitted_at":"2023-11-26T17:44:29Z","abstract_excerpt":"This study focuses on a Wasserstein-type gradient flow, which represents an optimization process of a continuous model of a Deep Neural Network (DNN). First, we establish the existence of a minimizer for an average loss of the model under $L^2$-regularization. Subsequently, we show the existence of a curve of maximal slope of the loss. Our main result is the convergence of flow to a critical point of the loss as time goes to infinity. An essential aspect of proving this result involves the establishment of the \\L{}ojasiewicz--Simon gradient inequality for the loss. We derive this inequality by"},"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":"2311.15365","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2023-11-26T17:44:29Z","cross_cats_sorted":["math.AP","math.FA","math.PR"],"title_canon_sha256":"bbcf647e67af344e59182e2ea0dd5d4100556a4e73bf44244b35c33eee880627","abstract_canon_sha256":"29f6d5933838004e8c27d3a12e2b296b9079ea3a30bc425d14b69244c99f70d3"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:07:31.060746Z","signature_b64":"Q9SCIhIFb62AzqoTTlTkEXEnk7C2+qmZjkWImRnJk1Rt8aULPqhFeFE7QL7ELlP7B3ZyQmb2/gMLid0rjzcgDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"a8a77aaa67a1dede6a67c2f48e65ac9a83270d171075537bb9947ed699f8dbd5","last_reissued_at":"2026-07-05T08:07:31.060250Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:07:31.060250Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"A convergence result of a continuous model of deep learning via \\L{}ojasiewicz--Simon inequality","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["math.AP","math.FA","math.PR"],"primary_cat":"cs.LG","authors_text":"Noboru Isobe","submitted_at":"2023-11-26T17:44:29Z","abstract_excerpt":"This study focuses on a Wasserstein-type gradient flow, which represents an optimization process of a continuous model of a Deep Neural Network (DNN). First, we establish the existence of a minimizer for an average loss of the model under $L^2$-regularization. Subsequently, we show the existence of a curve of maximal slope of the loss. Our main result is the convergence of flow to a critical point of the loss as time goes to infinity. An essential aspect of proving this result involves the establishment of the \\L{}ojasiewicz--Simon gradient inequality for the loss. We derive this inequality by"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2311.15365","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/2311.15365/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":"2311.15365","created_at":"2026-07-05T08:07:31.060306+00:00"},{"alias_kind":"arxiv_version","alias_value":"2311.15365v2","created_at":"2026-07-05T08:07:31.060306+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2311.15365","created_at":"2026-07-05T08:07:31.060306+00:00"},{"alias_kind":"pith_short_12","alias_value":"VCTXVKTHUHPN","created_at":"2026-07-05T08:07:31.060306+00:00"},{"alias_kind":"pith_short_16","alias_value":"VCTXVKTHUHPN42TH","created_at":"2026-07-05T08:07:31.060306+00:00"},{"alias_kind":"pith_short_8","alias_value":"VCTXVKTH","created_at":"2026-07-05T08:07:31.060306+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":3,"internal_anchor_count":3,"sample":[{"citing_arxiv_id":"2605.17660","citing_title":"Training Infinitely Deep and Wide Transformers","ref_index":27,"is_internal_anchor":true},{"citing_arxiv_id":"2605.11059","citing_title":"Uniform Scaling Limits in AdamW-Trained Transformers","ref_index":38,"is_internal_anchor":true},{"citing_arxiv_id":"2604.27052","citing_title":"Man, Machine, and Mathematics","ref_index":46,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/VCTXVKTHUHPN42THYL2I4ZNMTK","json":"https://pith.science/pith/VCTXVKTHUHPN42THYL2I4ZNMTK.json","graph_json":"https://pith.science/api/pith-number/VCTXVKTHUHPN42THYL2I4ZNMTK/graph.json","events_json":"https://pith.science/api/pith-number/VCTXVKTHUHPN42THYL2I4ZNMTK/events.json","paper":"https://pith.science/paper/VCTXVKTH"},"agent_actions":{"view_html":"https://pith.science/pith/VCTXVKTHUHPN42THYL2I4ZNMTK","download_json":"https://pith.science/pith/VCTXVKTHUHPN42THYL2I4ZNMTK.json","view_paper":"https://pith.science/paper/VCTXVKTH","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2311.15365&json=true","fetch_graph":"https://pith.science/api/pith-number/VCTXVKTHUHPN42THYL2I4ZNMTK/graph.json","fetch_events":"https://pith.science/api/pith-number/VCTXVKTHUHPN42THYL2I4ZNMTK/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/VCTXVKTHUHPN42THYL2I4ZNMTK/action/timestamp_anchor","attest_storage":"https://pith.science/pith/VCTXVKTHUHPN42THYL2I4ZNMTK/action/storage_attestation","attest_author":"https://pith.science/pith/VCTXVKTHUHPN42THYL2I4ZNMTK/action/author_attestation","sign_citation":"https://pith.science/pith/VCTXVKTHUHPN42THYL2I4ZNMTK/action/citation_signature","submit_replication":"https://pith.science/pith/VCTXVKTHUHPN42THYL2I4ZNMTK/action/replication_record"}},"created_at":"2026-07-05T08:07:31.060306+00:00","updated_at":"2026-07-05T08:07:31.060306+00:00"}