{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2021:XXJBPSGG25634V6POF3XTSV36M","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"2e11a8c68a9cb40a804cd0628c239c003dfcbf9ee36144a81b40bf717c0debaf","cross_cats_sorted":["math.OC"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2021-08-04T13:10:30Z","title_canon_sha256":"413165efe0e78318f1cb842bbad1f30314c322788b571f262ec4ef2a08d78152"},"schema_version":"1.0","source":{"id":"2108.02040","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2108.02040","created_at":"2026-07-05T03:34:45Z"},{"alias_kind":"arxiv_version","alias_value":"2108.02040v2","created_at":"2026-07-05T03:34:45Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2108.02040","created_at":"2026-07-05T03:34:45Z"},{"alias_kind":"pith_short_12","alias_value":"XXJBPSGG2563","created_at":"2026-07-05T03:34:45Z"},{"alias_kind":"pith_short_16","alias_value":"XXJBPSGG25634V6P","created_at":"2026-07-05T03:34:45Z"},{"alias_kind":"pith_short_8","alias_value":"XXJBPSGG","created_at":"2026-07-05T03:34:45Z"}],"graph_snapshots":[{"event_id":"sha256:9e13829b79dd27becfb3c65f9bf36b4b36a5cfb5ae93fe75efd3d2f9d6f53439","target":"graph","created_at":"2026-07-05T03:34:45Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2108.02040/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"We study the convergence properties of gradient descent for training deep linear neural networks, i.e., deep matrix factorizations, by extending a previous analysis for the related gradient flow. We show that under suitable conditions on the step sizes gradient descent converges to a critical point of the loss function, i.e., the square loss in this article. Furthermore, we demonstrate that for almost all initializations gradient descent converges to a global minimum in the case of two layers. In the case of three or more layers we show that gradient descent converges to a global minimum on th","authors_text":"Gabin Maxime Nguegnang, Holger Rauhut, Ulrich Terstiege","cross_cats":["math.OC"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2021-08-04T13:10:30Z","title":"Convergence of gradient descent for learning linear neural networks"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2108.02040","kind":"arxiv","version":2},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:4534c85d84b04f4e69e6105fa4981db2783d3dbd4a56e4f265ad0203ba5c0fab","target":"record","created_at":"2026-07-05T03:34:45Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"2e11a8c68a9cb40a804cd0628c239c003dfcbf9ee36144a81b40bf717c0debaf","cross_cats_sorted":["math.OC"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2021-08-04T13:10:30Z","title_canon_sha256":"413165efe0e78318f1cb842bbad1f30314c322788b571f262ec4ef2a08d78152"},"schema_version":"1.0","source":{"id":"2108.02040","kind":"arxiv","version":2}},"canonical_sha256":"bdd217c8c6d77dbe57cf717779cabbf3350b7af09a4c2e6692ad1dd1321c8c97","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"bdd217c8c6d77dbe57cf717779cabbf3350b7af09a4c2e6692ad1dd1321c8c97","first_computed_at":"2026-07-05T03:34:45.923659Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T03:34:45.923659Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"8e+lE3WQ+Phwo2fGzpcgRw0k2zrmcGz3Hc8FmUyM8S3FVBlDfryeleLUudnm3EAFYvSlpFLOHqOYf0k2ZyzXAA==","signature_status":"signed_v1","signed_at":"2026-07-05T03:34:45.924080Z","signed_message":"canonical_sha256_bytes"},"source_id":"2108.02040","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:4534c85d84b04f4e69e6105fa4981db2783d3dbd4a56e4f265ad0203ba5c0fab","sha256:9e13829b79dd27becfb3c65f9bf36b4b36a5cfb5ae93fe75efd3d2f9d6f53439"],"state_sha256":"b47c3861f1cad5dc34573ea18e47d93e185bdf13b87c7de8fd295103e801394e"}