{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2019:LAEUCRL2G7OGS7RMKFTL3P3LXG","short_pith_number":"pith:LAEUCRL2","canonical_record":{"source":{"id":"1901.08572","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2019-01-24T18:43:16Z","cross_cats_sorted":["stat.ML"],"title_canon_sha256":"8c17b45c1ee449286da7708e5b33a295ed164fc7b09bf443c614c743337de021","abstract_canon_sha256":"3a3b80f9264e0ed243568ef1b572f16296935ebfcf3d769e741d0bf83d96042c"},"schema_version":"1.0"},"canonical_sha256":"580941457a37dc697e2c5166bdbf6bb985498dc47874983ef95141166b85d033","source":{"kind":"arxiv","id":"1901.08572","version":3},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1901.08572","created_at":"2026-05-17T23:45:04Z"},{"alias_kind":"arxiv_version","alias_value":"1901.08572v3","created_at":"2026-05-17T23:45:04Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1901.08572","created_at":"2026-05-17T23:45:04Z"},{"alias_kind":"pith_short_12","alias_value":"LAEUCRL2G7OG","created_at":"2026-05-18T12:33:21Z"},{"alias_kind":"pith_short_16","alias_value":"LAEUCRL2G7OGS7RM","created_at":"2026-05-18T12:33:21Z"},{"alias_kind":"pith_short_8","alias_value":"LAEUCRL2","created_at":"2026-05-18T12:33:21Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2019:LAEUCRL2G7OGS7RMKFTL3P3LXG","target":"record","payload":{"canonical_record":{"source":{"id":"1901.08572","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2019-01-24T18:43:16Z","cross_cats_sorted":["stat.ML"],"title_canon_sha256":"8c17b45c1ee449286da7708e5b33a295ed164fc7b09bf443c614c743337de021","abstract_canon_sha256":"3a3b80f9264e0ed243568ef1b572f16296935ebfcf3d769e741d0bf83d96042c"},"schema_version":"1.0"},"canonical_sha256":"580941457a37dc697e2c5166bdbf6bb985498dc47874983ef95141166b85d033","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-17T23:45:04.073969Z","signature_b64":"tx2NzHy5fNzyOIAmS48gs4W2kc3422Mu/BrDezQDNNz04pjd3oVcf85OxR323KtPQZyoWJJJwYqhSg7Sl77BDg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"580941457a37dc697e2c5166bdbf6bb985498dc47874983ef95141166b85d033","last_reissued_at":"2026-05-17T23:45:04.073174Z","signature_status":"signed_v1","first_computed_at":"2026-05-17T23:45:04.073174Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"1901.08572","source_version":3,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-17T23:45:04Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"kAgxKuRduO5d7wu9RuZ0FN+/7xHIt9o1QVagMsh76u1P6LAtsjUzmdsMSHDhxdu/iP+FvlmWCegffDtgXIqGCg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-31T23:22:25.594862Z"},"content_sha256":"ea2f73c37179946519bca9f589e52d32e7d74803e9a2d561bc74addcb9f4f492","schema_version":"1.0","event_id":"sha256:ea2f73c37179946519bca9f589e52d32e7d74803e9a2d561bc74addcb9f4f492"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2019:LAEUCRL2G7OGS7RMKFTL3P3LXG","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Width Provably Matters in Optimization for Deep Linear Neural Networks","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Simon S. Du, Wei Hu","submitted_at":"2019-01-24T18:43:16Z","abstract_excerpt":"We prove that for an $L$-layer fully-connected linear neural network, if the width of every hidden layer is $\\tilde\\Omega (L \\cdot r \\cdot d_{\\mathrm{out}} \\cdot \\kappa^3 )$, where $r$ and $\\kappa$ are the rank and the condition number of the input data, and $d_{\\mathrm{out}}$ is the output dimension, then gradient descent with Gaussian random initialization converges to a global minimum at a linear rate. The number of iterations to find an $\\epsilon$-suboptimal solution is $O(\\kappa \\log(\\frac{1}{\\epsilon}))$. Our polynomial upper bound on the total running time for wide deep linear networks "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1901.08572","kind":"arxiv","version":3},"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"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-17T23:45:04Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"LdHugGRBSyJnOl4pf23PNmVeVTaE5Bn19MgMTWhfaJaR+TZEABHmM3sWzZbNzv/Xt+ofJ/hPS2OFdYZFlEByDg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-31T23:22:25.595601Z"},"content_sha256":"0b48eec6849efc0f40faf929ae2adfa8db00318be729532eb1d0db407f3b71c4","schema_version":"1.0","event_id":"sha256:0b48eec6849efc0f40faf929ae2adfa8db00318be729532eb1d0db407f3b71c4"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/LAEUCRL2G7OGS7RMKFTL3P3LXG/bundle.json","state_url":"https://pith.science/pith/LAEUCRL2G7OGS7RMKFTL3P3LXG/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/LAEUCRL2G7OGS7RMKFTL3P3LXG/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-05-31T23:22:25Z","links":{"resolver":"https://pith.science/pith/LAEUCRL2G7OGS7RMKFTL3P3LXG","bundle":"https://pith.science/pith/LAEUCRL2G7OGS7RMKFTL3P3LXG/bundle.json","state":"https://pith.science/pith/LAEUCRL2G7OGS7RMKFTL3P3LXG/state.json","well_known_bundle":"https://pith.science/.well-known/pith/LAEUCRL2G7OGS7RMKFTL3P3LXG/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2019:LAEUCRL2G7OGS7RMKFTL3P3LXG","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":"3a3b80f9264e0ed243568ef1b572f16296935ebfcf3d769e741d0bf83d96042c","cross_cats_sorted":["stat.ML"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2019-01-24T18:43:16Z","title_canon_sha256":"8c17b45c1ee449286da7708e5b33a295ed164fc7b09bf443c614c743337de021"},"schema_version":"1.0","source":{"id":"1901.08572","kind":"arxiv","version":3}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1901.08572","created_at":"2026-05-17T23:45:04Z"},{"alias_kind":"arxiv_version","alias_value":"1901.08572v3","created_at":"2026-05-17T23:45:04Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1901.08572","created_at":"2026-05-17T23:45:04Z"},{"alias_kind":"pith_short_12","alias_value":"LAEUCRL2G7OG","created_at":"2026-05-18T12:33:21Z"},{"alias_kind":"pith_short_16","alias_value":"LAEUCRL2G7OGS7RM","created_at":"2026-05-18T12:33:21Z"},{"alias_kind":"pith_short_8","alias_value":"LAEUCRL2","created_at":"2026-05-18T12:33:21Z"}],"graph_snapshots":[{"event_id":"sha256:0b48eec6849efc0f40faf929ae2adfa8db00318be729532eb1d0db407f3b71c4","target":"graph","created_at":"2026-05-17T23:45:04Z","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"},"paper":{"abstract_excerpt":"We prove that for an $L$-layer fully-connected linear neural network, if the width of every hidden layer is $\\tilde\\Omega (L \\cdot r \\cdot d_{\\mathrm{out}} \\cdot \\kappa^3 )$, where $r$ and $\\kappa$ are the rank and the condition number of the input data, and $d_{\\mathrm{out}}$ is the output dimension, then gradient descent with Gaussian random initialization converges to a global minimum at a linear rate. The number of iterations to find an $\\epsilon$-suboptimal solution is $O(\\kappa \\log(\\frac{1}{\\epsilon}))$. Our polynomial upper bound on the total running time for wide deep linear networks ","authors_text":"Simon S. Du, Wei Hu","cross_cats":["stat.ML"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2019-01-24T18:43:16Z","title":"Width Provably Matters in Optimization for Deep Linear Neural Networks"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1901.08572","kind":"arxiv","version":3},"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:ea2f73c37179946519bca9f589e52d32e7d74803e9a2d561bc74addcb9f4f492","target":"record","created_at":"2026-05-17T23:45:04Z","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":"3a3b80f9264e0ed243568ef1b572f16296935ebfcf3d769e741d0bf83d96042c","cross_cats_sorted":["stat.ML"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2019-01-24T18:43:16Z","title_canon_sha256":"8c17b45c1ee449286da7708e5b33a295ed164fc7b09bf443c614c743337de021"},"schema_version":"1.0","source":{"id":"1901.08572","kind":"arxiv","version":3}},"canonical_sha256":"580941457a37dc697e2c5166bdbf6bb985498dc47874983ef95141166b85d033","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"580941457a37dc697e2c5166bdbf6bb985498dc47874983ef95141166b85d033","first_computed_at":"2026-05-17T23:45:04.073174Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-17T23:45:04.073174Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"tx2NzHy5fNzyOIAmS48gs4W2kc3422Mu/BrDezQDNNz04pjd3oVcf85OxR323KtPQZyoWJJJwYqhSg7Sl77BDg==","signature_status":"signed_v1","signed_at":"2026-05-17T23:45:04.073969Z","signed_message":"canonical_sha256_bytes"},"source_id":"1901.08572","source_kind":"arxiv","source_version":3}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:ea2f73c37179946519bca9f589e52d32e7d74803e9a2d561bc74addcb9f4f492","sha256:0b48eec6849efc0f40faf929ae2adfa8db00318be729532eb1d0db407f3b71c4"],"state_sha256":"7be22b1f6af51dd3ed5c87b6c36ede7b33488ae5463ad9b155829ba989a1bc28"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"QfFRlXOqJi1H6QwV96Vzu7L6oC9Mcibd820iizCreylckhLdOSjBPhrH6oufjqtScKpxKlmeTyZl21kXuJPJAw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-05-31T23:22:25.599199Z","bundle_sha256":"4f5cbf6e328ed968f1256a57a641857dad3cd0ea8a49792cfbc71245d04d2917"}}