{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2019:LSW4UKSHY5ZHWJVQAGGZFJP5CM","short_pith_number":"pith:LSW4UKSH","canonical_record":{"source":{"id":"1903.11680","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2019-03-27T20:00:15Z","cross_cats_sorted":["stat.ML"],"title_canon_sha256":"a939426c238a2d04ffee6322a017c6d9d5b72d3c5a1f7dd2ddb87b08b7305e18","abstract_canon_sha256":"a0bf11e0b903233484ed69a71268535c8c685f767e5810e9721ccafbf50c47cf"},"schema_version":"1.0"},"canonical_sha256":"5cadca2a47c7727b26b0018d92a5fd1318ea459964c88663deed689a7661996f","source":{"kind":"arxiv","id":"1903.11680","version":3},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1903.11680","created_at":"2026-05-17T23:41:30Z"},{"alias_kind":"arxiv_version","alias_value":"1903.11680v3","created_at":"2026-05-17T23:41:30Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1903.11680","created_at":"2026-05-17T23:41:30Z"},{"alias_kind":"pith_short_12","alias_value":"LSW4UKSHY5ZH","created_at":"2026-05-18T12:33:21Z"},{"alias_kind":"pith_short_16","alias_value":"LSW4UKSHY5ZHWJVQ","created_at":"2026-05-18T12:33:21Z"},{"alias_kind":"pith_short_8","alias_value":"LSW4UKSH","created_at":"2026-05-18T12:33:21Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2019:LSW4UKSHY5ZHWJVQAGGZFJP5CM","target":"record","payload":{"canonical_record":{"source":{"id":"1903.11680","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2019-03-27T20:00:15Z","cross_cats_sorted":["stat.ML"],"title_canon_sha256":"a939426c238a2d04ffee6322a017c6d9d5b72d3c5a1f7dd2ddb87b08b7305e18","abstract_canon_sha256":"a0bf11e0b903233484ed69a71268535c8c685f767e5810e9721ccafbf50c47cf"},"schema_version":"1.0"},"canonical_sha256":"5cadca2a47c7727b26b0018d92a5fd1318ea459964c88663deed689a7661996f","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-17T23:41:30.518523Z","signature_b64":"R7pYa5UOv6g1GvxHZRelYll1g93jGRiccHXwZyffOQMCrYDOnrcG/ZMwrpWNMWBvkp5XmC7EWEp7rPvQCUVLDw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"5cadca2a47c7727b26b0018d92a5fd1318ea459964c88663deed689a7661996f","last_reissued_at":"2026-05-17T23:41:30.517746Z","signature_status":"signed_v1","first_computed_at":"2026-05-17T23:41:30.517746Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"1903.11680","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":"2026-05-17T23:41:30Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"t+h3UcEcdpx3sAlqasV5sb1SGnYm6EJjs7sAKC3GasHLlRKo+3J3Yu1XMsLSY54RzSlbLmIjJjoMToNppe8uCw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-26T13:36:06.631484Z"},"content_sha256":"c8cdb64d9de08d9b99379ec1d695602ef2cc84e4be1b379dbf45f9b33e9a9b0f","schema_version":"1.0","event_id":"sha256:c8cdb64d9de08d9b99379ec1d695602ef2cc84e4be1b379dbf45f9b33e9a9b0f"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2019:LSW4UKSHY5ZHWJVQAGGZFJP5CM","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Gradient Descent with Early Stopping is Provably Robust to Label Noise for Overparameterized Neural Networks","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Mahdi Soltanolkotabi, Mingchen Li, Samet Oymak","submitted_at":"2019-03-27T20:00:15Z","abstract_excerpt":"Modern neural networks are typically trained in an over-parameterized regime where the parameters of the model far exceed the size of the training data. Such neural networks in principle have the capacity to (over)fit any set of labels including pure noise. Despite this, somewhat paradoxically, neural network models trained via first-order methods continue to predict well on yet unseen test data. This paper takes a step towards demystifying this phenomena. Under a rich dataset model, we show that gradient descent is provably robust to noise/corruption on a constant fraction of the labels despi"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1903.11680","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":"2026-05-17T23:41:30Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"1VTCR31uABSfCCmxYJ7iz8XFylJz7uSs8uhT7OvCSW7diauyScps6XrRdJ/mYZl46PVFvGalyw8hNxHuOBKmDg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-26T13:36:06.632126Z"},"content_sha256":"49d25323c6d5e8f2429617b02de4c671c82e39710d827bbf90f68de0c6e31a94","schema_version":"1.0","event_id":"sha256:49d25323c6d5e8f2429617b02de4c671c82e39710d827bbf90f68de0c6e31a94"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/LSW4UKSHY5ZHWJVQAGGZFJP5CM/bundle.json","state_url":"https://pith.science/pith/LSW4UKSHY5ZHWJVQAGGZFJP5CM/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/LSW4UKSHY5ZHWJVQAGGZFJP5CM/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":"2026-05-26T13:36:06Z","links":{"resolver":"https://pith.science/pith/LSW4UKSHY5ZHWJVQAGGZFJP5CM","bundle":"https://pith.science/pith/LSW4UKSHY5ZHWJVQAGGZFJP5CM/bundle.json","state":"https://pith.science/pith/LSW4UKSHY5ZHWJVQAGGZFJP5CM/state.json","well_known_bundle":"https://pith.science/.well-known/pith/LSW4UKSHY5ZHWJVQAGGZFJP5CM/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2019:LSW4UKSHY5ZHWJVQAGGZFJP5CM","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":"a0bf11e0b903233484ed69a71268535c8c685f767e5810e9721ccafbf50c47cf","cross_cats_sorted":["stat.ML"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2019-03-27T20:00:15Z","title_canon_sha256":"a939426c238a2d04ffee6322a017c6d9d5b72d3c5a1f7dd2ddb87b08b7305e18"},"schema_version":"1.0","source":{"id":"1903.11680","kind":"arxiv","version":3}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1903.11680","created_at":"2026-05-17T23:41:30Z"},{"alias_kind":"arxiv_version","alias_value":"1903.11680v3","created_at":"2026-05-17T23:41:30Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1903.11680","created_at":"2026-05-17T23:41:30Z"},{"alias_kind":"pith_short_12","alias_value":"LSW4UKSHY5ZH","created_at":"2026-05-18T12:33:21Z"},{"alias_kind":"pith_short_16","alias_value":"LSW4UKSHY5ZHWJVQ","created_at":"2026-05-18T12:33:21Z"},{"alias_kind":"pith_short_8","alias_value":"LSW4UKSH","created_at":"2026-05-18T12:33:21Z"}],"graph_snapshots":[{"event_id":"sha256:49d25323c6d5e8f2429617b02de4c671c82e39710d827bbf90f68de0c6e31a94","target":"graph","created_at":"2026-05-17T23:41:30Z","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":"Modern neural networks are typically trained in an over-parameterized regime where the parameters of the model far exceed the size of the training data. Such neural networks in principle have the capacity to (over)fit any set of labels including pure noise. Despite this, somewhat paradoxically, neural network models trained via first-order methods continue to predict well on yet unseen test data. This paper takes a step towards demystifying this phenomena. Under a rich dataset model, we show that gradient descent is provably robust to noise/corruption on a constant fraction of the labels despi","authors_text":"Mahdi Soltanolkotabi, Mingchen Li, Samet Oymak","cross_cats":["stat.ML"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2019-03-27T20:00:15Z","title":"Gradient Descent with Early Stopping is Provably Robust to Label Noise for Overparameterized Neural Networks"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1903.11680","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:c8cdb64d9de08d9b99379ec1d695602ef2cc84e4be1b379dbf45f9b33e9a9b0f","target":"record","created_at":"2026-05-17T23:41:30Z","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":"a0bf11e0b903233484ed69a71268535c8c685f767e5810e9721ccafbf50c47cf","cross_cats_sorted":["stat.ML"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2019-03-27T20:00:15Z","title_canon_sha256":"a939426c238a2d04ffee6322a017c6d9d5b72d3c5a1f7dd2ddb87b08b7305e18"},"schema_version":"1.0","source":{"id":"1903.11680","kind":"arxiv","version":3}},"canonical_sha256":"5cadca2a47c7727b26b0018d92a5fd1318ea459964c88663deed689a7661996f","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"5cadca2a47c7727b26b0018d92a5fd1318ea459964c88663deed689a7661996f","first_computed_at":"2026-05-17T23:41:30.517746Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-17T23:41:30.517746Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"R7pYa5UOv6g1GvxHZRelYll1g93jGRiccHXwZyffOQMCrYDOnrcG/ZMwrpWNMWBvkp5XmC7EWEp7rPvQCUVLDw==","signature_status":"signed_v1","signed_at":"2026-05-17T23:41:30.518523Z","signed_message":"canonical_sha256_bytes"},"source_id":"1903.11680","source_kind":"arxiv","source_version":3}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:c8cdb64d9de08d9b99379ec1d695602ef2cc84e4be1b379dbf45f9b33e9a9b0f","sha256:49d25323c6d5e8f2429617b02de4c671c82e39710d827bbf90f68de0c6e31a94"],"state_sha256":"c323f0e44c9fab970871d75aabb28632d136e2efab5079c9c524cb806e8948cf"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"B47RCwpo+Evw60WP7wZGF2TIkroxAn5uHmabocv47xWg8iGvwS1QRE9uFoIlpDKdZJKUcJnAZNBKpoXUQ9LqDA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-05-26T13:36:06.635482Z","bundle_sha256":"16718ff97cceb51f0cb45be7ab5a2596dff3af00a3d1c31226ab9732bfd1e05c"}}