{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:AIMT6Q3B4XGID5SGETMOFWJF5I","short_pith_number":"pith:AIMT6Q3B","canonical_record":{"source":{"id":"2501.09137","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-01-15T20:43:36Z","cross_cats_sorted":["math.OC","stat.ML"],"title_canon_sha256":"708a197a15e0936e023b25226ce321890995eac9b99e19ed6d90de40ef2d8d2e","abstract_canon_sha256":"81a19dd9e901b00736c5e70afbbc68eb03d0fa896feed3406ba4b7efc0ac9302"},"schema_version":"1.0"},"canonical_sha256":"02193f4361e5cc81f64624d8e2d925ea339fc929407a0ae652e047be097abc9d","source":{"kind":"arxiv","id":"2501.09137","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2501.09137","created_at":"2026-07-05T10:02:58Z"},{"alias_kind":"arxiv_version","alias_value":"2501.09137v2","created_at":"2026-07-05T10:02:58Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2501.09137","created_at":"2026-07-05T10:02:58Z"},{"alias_kind":"pith_short_12","alias_value":"AIMT6Q3B4XGI","created_at":"2026-07-05T10:02:58Z"},{"alias_kind":"pith_short_16","alias_value":"AIMT6Q3B4XGID5SG","created_at":"2026-07-05T10:02:58Z"},{"alias_kind":"pith_short_8","alias_value":"AIMT6Q3B","created_at":"2026-07-05T10:02:58Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:AIMT6Q3B4XGID5SGETMOFWJF5I","target":"record","payload":{"canonical_record":{"source":{"id":"2501.09137","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-01-15T20:43:36Z","cross_cats_sorted":["math.OC","stat.ML"],"title_canon_sha256":"708a197a15e0936e023b25226ce321890995eac9b99e19ed6d90de40ef2d8d2e","abstract_canon_sha256":"81a19dd9e901b00736c5e70afbbc68eb03d0fa896feed3406ba4b7efc0ac9302"},"schema_version":"1.0"},"canonical_sha256":"02193f4361e5cc81f64624d8e2d925ea339fc929407a0ae652e047be097abc9d","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:02:58.617640Z","signature_b64":"I9YMyPKRALCZPzQnmbLzHj+4O5+e9U/L8jbsKZD6RYXXRFFAUbC46Yf7nMEWMoiO3dMF4cQxWCRiIszNkUe/Bg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"02193f4361e5cc81f64624d8e2d925ea339fc929407a0ae652e047be097abc9d","last_reissued_at":"2026-07-05T10:02:58.617146Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:02:58.617146Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2501.09137","source_version":2,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T10:02:58Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"IEZRAvshHDoJcv6vxpwyJ2tSoj/FJXU/RCn0DU53qV4UtWankpziwMYFKdHLTYM3xTmb+sE4OndJa4cBR9UbAw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-22T13:40:15.015713Z"},"content_sha256":"60e9fa891dff9d8962cabc16224797835c3f45416e677ed5b31792e9794c54f7","schema_version":"1.0","event_id":"sha256:60e9fa891dff9d8962cabc16224797835c3f45416e677ed5b31792e9794c54f7"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:AIMT6Q3B4XGID5SGETMOFWJF5I","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Gradient Descent Converges Linearly to Flatter Minima than Gradient Flow in Shallow Linear Networks","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["math.OC","stat.ML"],"primary_cat":"cs.LG","authors_text":"Blake Woodworth, Pierfrancesco Beneventano","submitted_at":"2025-01-15T20:43:36Z","abstract_excerpt":"We study the gradient descent (GD) dynamics of a depth-2 linear neural network with a single input and output. We show that GD converges at an explicit linear rate to a global minimum of the training loss, even with a large stepsize -- about $2/\\textrm{sharpness}$. It still converges for even larger stepsizes, but may do so very slowly. We also characterize the solution to which GD converges, which has lower norm and sharpness than the gradient flow solution. Our analysis reveals a trade off between the speed of convergence and the magnitude of implicit regularization. This sheds light on the "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2501.09137","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/2501.09137/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"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T10:02:58Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"G1scCrYepXfQqtvH4XIEX92kb78gxWvcaCnQvbFOFXukgnejeLS5vDTfD2wM8ADDUm+8I+JnFVxhSpm86d8wBg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-22T13:40:15.016109Z"},"content_sha256":"ae3cae9d0c619968e06b3907d40ffa7935a5b4628ddfeabbe47cd43e63cc06b0","schema_version":"1.0","event_id":"sha256:ae3cae9d0c619968e06b3907d40ffa7935a5b4628ddfeabbe47cd43e63cc06b0"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/AIMT6Q3B4XGID5SGETMOFWJF5I/bundle.json","state_url":"https://pith.science/pith/AIMT6Q3B4XGID5SGETMOFWJF5I/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/AIMT6Q3B4XGID5SGETMOFWJF5I/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":"2026-07-22T13:40:15Z","links":{"resolver":"https://pith.science/pith/AIMT6Q3B4XGID5SGETMOFWJF5I","bundle":"https://pith.science/pith/AIMT6Q3B4XGID5SGETMOFWJF5I/bundle.json","state":"https://pith.science/pith/AIMT6Q3B4XGID5SGETMOFWJF5I/state.json","well_known_bundle":"https://pith.science/.well-known/pith/AIMT6Q3B4XGID5SGETMOFWJF5I/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:AIMT6Q3B4XGID5SGETMOFWJF5I","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":"81a19dd9e901b00736c5e70afbbc68eb03d0fa896feed3406ba4b7efc0ac9302","cross_cats_sorted":["math.OC","stat.ML"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-01-15T20:43:36Z","title_canon_sha256":"708a197a15e0936e023b25226ce321890995eac9b99e19ed6d90de40ef2d8d2e"},"schema_version":"1.0","source":{"id":"2501.09137","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2501.09137","created_at":"2026-07-05T10:02:58Z"},{"alias_kind":"arxiv_version","alias_value":"2501.09137v2","created_at":"2026-07-05T10:02:58Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2501.09137","created_at":"2026-07-05T10:02:58Z"},{"alias_kind":"pith_short_12","alias_value":"AIMT6Q3B4XGI","created_at":"2026-07-05T10:02:58Z"},{"alias_kind":"pith_short_16","alias_value":"AIMT6Q3B4XGID5SG","created_at":"2026-07-05T10:02:58Z"},{"alias_kind":"pith_short_8","alias_value":"AIMT6Q3B","created_at":"2026-07-05T10:02:58Z"}],"graph_snapshots":[{"event_id":"sha256:ae3cae9d0c619968e06b3907d40ffa7935a5b4628ddfeabbe47cd43e63cc06b0","target":"graph","created_at":"2026-07-05T10:02:58Z","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/2501.09137/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"We study the gradient descent (GD) dynamics of a depth-2 linear neural network with a single input and output. We show that GD converges at an explicit linear rate to a global minimum of the training loss, even with a large stepsize -- about $2/\\textrm{sharpness}$. It still converges for even larger stepsizes, but may do so very slowly. We also characterize the solution to which GD converges, which has lower norm and sharpness than the gradient flow solution. Our analysis reveals a trade off between the speed of convergence and the magnitude of implicit regularization. This sheds light on the ","authors_text":"Blake Woodworth, Pierfrancesco Beneventano","cross_cats":["math.OC","stat.ML"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-01-15T20:43:36Z","title":"Gradient Descent Converges Linearly to Flatter Minima than Gradient Flow in Shallow Linear Networks"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2501.09137","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":"sha256:60e9fa891dff9d8962cabc16224797835c3f45416e677ed5b31792e9794c54f7","target":"record","created_at":"2026-07-05T10:02:58Z","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":"81a19dd9e901b00736c5e70afbbc68eb03d0fa896feed3406ba4b7efc0ac9302","cross_cats_sorted":["math.OC","stat.ML"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-01-15T20:43:36Z","title_canon_sha256":"708a197a15e0936e023b25226ce321890995eac9b99e19ed6d90de40ef2d8d2e"},"schema_version":"1.0","source":{"id":"2501.09137","kind":"arxiv","version":2}},"canonical_sha256":"02193f4361e5cc81f64624d8e2d925ea339fc929407a0ae652e047be097abc9d","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"02193f4361e5cc81f64624d8e2d925ea339fc929407a0ae652e047be097abc9d","first_computed_at":"2026-07-05T10:02:58.617146Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T10:02:58.617146Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"I9YMyPKRALCZPzQnmbLzHj+4O5+e9U/L8jbsKZD6RYXXRFFAUbC46Yf7nMEWMoiO3dMF4cQxWCRiIszNkUe/Bg==","signature_status":"signed_v1","signed_at":"2026-07-05T10:02:58.617640Z","signed_message":"canonical_sha256_bytes"},"source_id":"2501.09137","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:60e9fa891dff9d8962cabc16224797835c3f45416e677ed5b31792e9794c54f7","sha256:ae3cae9d0c619968e06b3907d40ffa7935a5b4628ddfeabbe47cd43e63cc06b0"],"state_sha256":"1da61068a8ec2612cd9c33b0a49a80557f722e04b11c5971fe3f756f7d5e2de1"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"lD+OuNpZbI3KoGORDiP7DwNMa456ZfjNzCFZSIDKFwH3iGxqMQh4Kt/efNec1+Q2eUN1dem75pRKhn6gYcLXCg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-07-22T13:40:15.018442Z","bundle_sha256":"54e0f3c08add52812e3ad0e9c7d366d1b70fec9118619eff3a8fee43166afd75"}}