{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:UJRDM2WDDVUZOFLNN4SIPLDGDD","short_pith_number":"pith:UJRDM2WD","schema_version":"1.0","canonical_sha256":"a262366ac31d6997156d6f2487ac6618f88d325fee23110e87895b9892766afc","source":{"kind":"arxiv","id":"2107.07572","version":2},"attestation_state":"computed","paper":{"title":"Globally Convergent Multilevel Training of Deep Residual Networks","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":["cs.NA","math.NA","math.OC"],"primary_cat":"cs.LG","authors_text":"Alena Kopani\\v{c}\\'akov\\'a, Rolf Krause","submitted_at":"2021-07-15T19:08:58Z","abstract_excerpt":"We propose a globally convergent multilevel training method for deep residual networks (ResNets). The devised method can be seen as a novel variant of the recursive multilevel trust-region (RMTR) method, which operates in hybrid (stochastic-deterministic) settings by adaptively adjusting mini-batch sizes during the training. The multilevel hierarchy and the transfer operators are constructed by exploiting a dynamical system's viewpoint, which interprets forward propagation through the ResNet as a forward Euler discretization of an initial value problem. In contrast to traditional training appr"},"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":"2107.07572","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.LG","submitted_at":"2021-07-15T19:08:58Z","cross_cats_sorted":["cs.NA","math.NA","math.OC"],"title_canon_sha256":"9e5b23967091d11d58906ae9ace5da3bde0945b9d9f5dc1dbe1d338724262dd6","abstract_canon_sha256":"d6ab32ba39699653010c77c3041e4d26d1d098a258b21f3ffe8fdba32f95dc80"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T04:30:51.838393Z","signature_b64":"Cvoo86hrABgIF5oVXnXKkc1V85jo0cqw8O9R7u9CETfluPQXaggkFCQ3OTHXB+tJ1CvIrojFS86ggapzzeRRDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"a262366ac31d6997156d6f2487ac6618f88d325fee23110e87895b9892766afc","last_reissued_at":"2026-07-05T04:30:51.838004Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T04:30:51.838004Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Globally Convergent Multilevel Training of Deep Residual Networks","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":["cs.NA","math.NA","math.OC"],"primary_cat":"cs.LG","authors_text":"Alena Kopani\\v{c}\\'akov\\'a, Rolf Krause","submitted_at":"2021-07-15T19:08:58Z","abstract_excerpt":"We propose a globally convergent multilevel training method for deep residual networks (ResNets). The devised method can be seen as a novel variant of the recursive multilevel trust-region (RMTR) method, which operates in hybrid (stochastic-deterministic) settings by adaptively adjusting mini-batch sizes during the training. The multilevel hierarchy and the transfer operators are constructed by exploiting a dynamical system's viewpoint, which interprets forward propagation through the ResNet as a forward Euler discretization of an initial value problem. In contrast to traditional training appr"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2107.07572","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/2107.07572/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":"2107.07572","created_at":"2026-07-05T04:30:51.838059+00:00"},{"alias_kind":"arxiv_version","alias_value":"2107.07572v2","created_at":"2026-07-05T04:30:51.838059+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2107.07572","created_at":"2026-07-05T04:30:51.838059+00:00"},{"alias_kind":"pith_short_12","alias_value":"UJRDM2WDDVUZ","created_at":"2026-07-05T04:30:51.838059+00:00"},{"alias_kind":"pith_short_16","alias_value":"UJRDM2WDDVUZOFLN","created_at":"2026-07-05T04:30:51.838059+00:00"},{"alias_kind":"pith_short_8","alias_value":"UJRDM2WD","created_at":"2026-07-05T04:30:51.838059+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/UJRDM2WDDVUZOFLNN4SIPLDGDD","json":"https://pith.science/pith/UJRDM2WDDVUZOFLNN4SIPLDGDD.json","graph_json":"https://pith.science/api/pith-number/UJRDM2WDDVUZOFLNN4SIPLDGDD/graph.json","events_json":"https://pith.science/api/pith-number/UJRDM2WDDVUZOFLNN4SIPLDGDD/events.json","paper":"https://pith.science/paper/UJRDM2WD"},"agent_actions":{"view_html":"https://pith.science/pith/UJRDM2WDDVUZOFLNN4SIPLDGDD","download_json":"https://pith.science/pith/UJRDM2WDDVUZOFLNN4SIPLDGDD.json","view_paper":"https://pith.science/paper/UJRDM2WD","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2107.07572&json=true","fetch_graph":"https://pith.science/api/pith-number/UJRDM2WDDVUZOFLNN4SIPLDGDD/graph.json","fetch_events":"https://pith.science/api/pith-number/UJRDM2WDDVUZOFLNN4SIPLDGDD/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/UJRDM2WDDVUZOFLNN4SIPLDGDD/action/timestamp_anchor","attest_storage":"https://pith.science/pith/UJRDM2WDDVUZOFLNN4SIPLDGDD/action/storage_attestation","attest_author":"https://pith.science/pith/UJRDM2WDDVUZOFLNN4SIPLDGDD/action/author_attestation","sign_citation":"https://pith.science/pith/UJRDM2WDDVUZOFLNN4SIPLDGDD/action/citation_signature","submit_replication":"https://pith.science/pith/UJRDM2WDDVUZOFLNN4SIPLDGDD/action/replication_record"}},"created_at":"2026-07-05T04:30:51.838059+00:00","updated_at":"2026-07-05T04:30:51.838059+00:00"}