{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:HCOQHVM5EKZAWPMPLT46VR4LCW","short_pith_number":"pith:HCOQHVM5","schema_version":"1.0","canonical_sha256":"389d03d59d22b20b3d8f5cf9eac78b15b93a90595f22bba7314add30f617d773","source":{"kind":"arxiv","id":"2605.14860","version":1},"attestation_state":"computed","paper":{"title":"A Non-Monotone Preconditioned Trust-Region Method for Neural Network Training","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"math.OC","authors_text":"Andrea Angino, Bindi \\c{C}apriqi, Ken Trotti, Rolf Krause, Shega Likaj","submitted_at":"2026-05-14T14:06:51Z","abstract_excerpt":"Training deep neural networks at scale can benefit from domain decomposition, where the network is split into subdomains trained in parallel and coupled by a global trust-region mechanism. Building on the Additively Preconditioned Trust-Region Strategy (APTS), we propose a non-monotone variant with a nonlinear additive Schwarz preconditioner that combines parallel subdomain corrections with global coarse-space directions. A windowed acceptance criterion allows controlled objective increases, avoiding needless rejection of effective coarse steps. The resulting non-monotone APTS (NAPTS) preserve"},"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":"2605.14860","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"math.OC","submitted_at":"2026-05-14T14:06:51Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"cab08fc0e15956dea93828e4de7afddcd8468e4577198ea2b6f26bfb137cd422","abstract_canon_sha256":"b1f1da2106c703792a61e448a5b04d5d47f75571d5c3994d87d656b13547b708"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-17T23:38:56.241978Z","signature_b64":"jxw3LCXItlom/sL/Hv56WDUNr+uRiQTE3NX4yZ4eV7VQcggiPnv7+bxQ7hQSWQrsrEG5r7SuXa2CaxZ+MGUDBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"389d03d59d22b20b3d8f5cf9eac78b15b93a90595f22bba7314add30f617d773","last_reissued_at":"2026-05-17T23:38:56.241282Z","signature_status":"signed_v1","first_computed_at":"2026-05-17T23:38:56.241282Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"A Non-Monotone Preconditioned Trust-Region Method for Neural Network Training","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"math.OC","authors_text":"Andrea Angino, Bindi \\c{C}apriqi, Ken Trotti, Rolf Krause, Shega Likaj","submitted_at":"2026-05-14T14:06:51Z","abstract_excerpt":"Training deep neural networks at scale can benefit from domain decomposition, where the network is split into subdomains trained in parallel and coupled by a global trust-region mechanism. Building on the Additively Preconditioned Trust-Region Strategy (APTS), we propose a non-monotone variant with a nonlinear additive Schwarz preconditioner that combines parallel subdomain corrections with global coarse-space directions. A windowed acceptance criterion allows controlled objective increases, avoiding needless rejection of effective coarse steps. The resulting non-monotone APTS (NAPTS) preserve"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.14860","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":"2605.14860","created_at":"2026-05-17T23:38:56.241406+00:00"},{"alias_kind":"arxiv_version","alias_value":"2605.14860v1","created_at":"2026-05-17T23:38:56.241406+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.14860","created_at":"2026-05-17T23:38:56.241406+00:00"},{"alias_kind":"pith_short_12","alias_value":"HCOQHVM5EKZA","created_at":"2026-05-18T12:33:37.589309+00:00"},{"alias_kind":"pith_short_16","alias_value":"HCOQHVM5EKZAWPMP","created_at":"2026-05-18T12:33:37.589309+00:00"},{"alias_kind":"pith_short_8","alias_value":"HCOQHVM5","created_at":"2026-05-18T12:33:37.589309+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/HCOQHVM5EKZAWPMPLT46VR4LCW","json":"https://pith.science/pith/HCOQHVM5EKZAWPMPLT46VR4LCW.json","graph_json":"https://pith.science/api/pith-number/HCOQHVM5EKZAWPMPLT46VR4LCW/graph.json","events_json":"https://pith.science/api/pith-number/HCOQHVM5EKZAWPMPLT46VR4LCW/events.json","paper":"https://pith.science/paper/HCOQHVM5"},"agent_actions":{"view_html":"https://pith.science/pith/HCOQHVM5EKZAWPMPLT46VR4LCW","download_json":"https://pith.science/pith/HCOQHVM5EKZAWPMPLT46VR4LCW.json","view_paper":"https://pith.science/paper/HCOQHVM5","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2605.14860&json=true","fetch_graph":"https://pith.science/api/pith-number/HCOQHVM5EKZAWPMPLT46VR4LCW/graph.json","fetch_events":"https://pith.science/api/pith-number/HCOQHVM5EKZAWPMPLT46VR4LCW/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/HCOQHVM5EKZAWPMPLT46VR4LCW/action/timestamp_anchor","attest_storage":"https://pith.science/pith/HCOQHVM5EKZAWPMPLT46VR4LCW/action/storage_attestation","attest_author":"https://pith.science/pith/HCOQHVM5EKZAWPMPLT46VR4LCW/action/author_attestation","sign_citation":"https://pith.science/pith/HCOQHVM5EKZAWPMPLT46VR4LCW/action/citation_signature","submit_replication":"https://pith.science/pith/HCOQHVM5EKZAWPMPLT46VR4LCW/action/replication_record"}},"created_at":"2026-05-17T23:38:56.241406+00:00","updated_at":"2026-05-17T23:38:56.241406+00:00"}