{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2018:MJ2WA57DQNU52CGKLQGUREQPWC","short_pith_number":"pith:MJ2WA57D","canonical_record":{"source":{"id":"1811.08888","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2018-11-21T18:58:46Z","cross_cats_sorted":["cs.AI","math.OC","stat.ML"],"title_canon_sha256":"ba3d66e38335314c46352c85f975f0a56f2edf99fa82881beb3d59975108a5c7","abstract_canon_sha256":"9b3ac1e631028645c4f2f77f7b1033aa2429571b6f257d3d903a7b2e6a28e50f"},"schema_version":"1.0"},"canonical_sha256":"62756077e38369dd08ca5c0d48920fb08e08ea73a30ffc793368bcb2808112f5","source":{"kind":"arxiv","id":"1811.08888","version":3},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1811.08888","created_at":"2026-05-17T23:57:21Z"},{"alias_kind":"arxiv_version","alias_value":"1811.08888v3","created_at":"2026-05-17T23:57:21Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1811.08888","created_at":"2026-05-17T23:57:21Z"},{"alias_kind":"pith_short_12","alias_value":"MJ2WA57DQNU5","created_at":"2026-05-18T12:32:37Z"},{"alias_kind":"pith_short_16","alias_value":"MJ2WA57DQNU52CGK","created_at":"2026-05-18T12:32:37Z"},{"alias_kind":"pith_short_8","alias_value":"MJ2WA57D","created_at":"2026-05-18T12:32:37Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2018:MJ2WA57DQNU52CGKLQGUREQPWC","target":"record","payload":{"canonical_record":{"source":{"id":"1811.08888","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2018-11-21T18:58:46Z","cross_cats_sorted":["cs.AI","math.OC","stat.ML"],"title_canon_sha256":"ba3d66e38335314c46352c85f975f0a56f2edf99fa82881beb3d59975108a5c7","abstract_canon_sha256":"9b3ac1e631028645c4f2f77f7b1033aa2429571b6f257d3d903a7b2e6a28e50f"},"schema_version":"1.0"},"canonical_sha256":"62756077e38369dd08ca5c0d48920fb08e08ea73a30ffc793368bcb2808112f5","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-17T23:57:21.181773Z","signature_b64":"9rGZulk9NOyfgjf3SYJdn2Glu6yZLbaYVcifoKVKzM2LNgYhW+l8vUUq5X5mxJaCjk+c37LB8ZXGIaauX8GcAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"62756077e38369dd08ca5c0d48920fb08e08ea73a30ffc793368bcb2808112f5","last_reissued_at":"2026-05-17T23:57:21.181083Z","signature_status":"signed_v1","first_computed_at":"2026-05-17T23:57:21.181083Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"1811.08888","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":"2026-05-17T23:57:21Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"zm/QTLN/9KL4k5SDd3PccOJLOHot0+PLe0iHm3uGiNVtpxg1NWZCjl60XYKC2yAPers4BCnMdLRuONtA1nqmAQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-28T21:45:53.384558Z"},"content_sha256":"99796a43c450e8636dffe4f25f75c5ed3da6bb0857fc4abaa354714312e37ca3","schema_version":"1.0","event_id":"sha256:99796a43c450e8636dffe4f25f75c5ed3da6bb0857fc4abaa354714312e37ca3"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2018:MJ2WA57DQNU52CGKLQGUREQPWC","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Stochastic Gradient Descent Optimizes Over-parameterized Deep ReLU Networks","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","math.OC","stat.ML"],"primary_cat":"cs.LG","authors_text":"Difan Zou, Dongruo Zhou, Quanquan Gu, Yuan Cao","submitted_at":"2018-11-21T18:58:46Z","abstract_excerpt":"We study the problem of training deep neural networks with Rectified Linear Unit (ReLU) activation function using gradient descent and stochastic gradient descent. In particular, we study the binary classification problem and show that for a broad family of loss functions, with proper random weight initialization, both gradient descent and stochastic gradient descent can find the global minima of the training loss for an over-parameterized deep ReLU network, under mild assumption on the training data. The key idea of our proof is that Gaussian random initialization followed by (stochastic) gra"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1811.08888","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":"2026-05-17T23:57:21Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"30aNi9aE4CLF0alV6RFjEvSFizie3zT1civnbbNsuRQFeOVn1anEk32ZmCJT8Rz+mKsZmns1+ctylmOeR8oTAQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-28T21:45:53.384920Z"},"content_sha256":"945b8a181061d8160da1ba4a3074de1509df74a227df5835f3c7b4cbc4657976","schema_version":"1.0","event_id":"sha256:945b8a181061d8160da1ba4a3074de1509df74a227df5835f3c7b4cbc4657976"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/MJ2WA57DQNU52CGKLQGUREQPWC/bundle.json","state_url":"https://pith.science/pith/MJ2WA57DQNU52CGKLQGUREQPWC/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/MJ2WA57DQNU52CGKLQGUREQPWC/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":"2026-05-28T21:45:53Z","links":{"resolver":"https://pith.science/pith/MJ2WA57DQNU52CGKLQGUREQPWC","bundle":"https://pith.science/pith/MJ2WA57DQNU52CGKLQGUREQPWC/bundle.json","state":"https://pith.science/pith/MJ2WA57DQNU52CGKLQGUREQPWC/state.json","well_known_bundle":"https://pith.science/.well-known/pith/MJ2WA57DQNU52CGKLQGUREQPWC/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2018:MJ2WA57DQNU52CGKLQGUREQPWC","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":"9b3ac1e631028645c4f2f77f7b1033aa2429571b6f257d3d903a7b2e6a28e50f","cross_cats_sorted":["cs.AI","math.OC","stat.ML"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2018-11-21T18:58:46Z","title_canon_sha256":"ba3d66e38335314c46352c85f975f0a56f2edf99fa82881beb3d59975108a5c7"},"schema_version":"1.0","source":{"id":"1811.08888","kind":"arxiv","version":3}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1811.08888","created_at":"2026-05-17T23:57:21Z"},{"alias_kind":"arxiv_version","alias_value":"1811.08888v3","created_at":"2026-05-17T23:57:21Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1811.08888","created_at":"2026-05-17T23:57:21Z"},{"alias_kind":"pith_short_12","alias_value":"MJ2WA57DQNU5","created_at":"2026-05-18T12:32:37Z"},{"alias_kind":"pith_short_16","alias_value":"MJ2WA57DQNU52CGK","created_at":"2026-05-18T12:32:37Z"},{"alias_kind":"pith_short_8","alias_value":"MJ2WA57D","created_at":"2026-05-18T12:32:37Z"}],"graph_snapshots":[{"event_id":"sha256:945b8a181061d8160da1ba4a3074de1509df74a227df5835f3c7b4cbc4657976","target":"graph","created_at":"2026-05-17T23:57:21Z","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 study the problem of training deep neural networks with Rectified Linear Unit (ReLU) activation function using gradient descent and stochastic gradient descent. In particular, we study the binary classification problem and show that for a broad family of loss functions, with proper random weight initialization, both gradient descent and stochastic gradient descent can find the global minima of the training loss for an over-parameterized deep ReLU network, under mild assumption on the training data. The key idea of our proof is that Gaussian random initialization followed by (stochastic) gra","authors_text":"Difan Zou, Dongruo Zhou, Quanquan Gu, Yuan Cao","cross_cats":["cs.AI","math.OC","stat.ML"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2018-11-21T18:58:46Z","title":"Stochastic Gradient Descent Optimizes Over-parameterized Deep ReLU Networks"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1811.08888","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":"sha256:99796a43c450e8636dffe4f25f75c5ed3da6bb0857fc4abaa354714312e37ca3","target":"record","created_at":"2026-05-17T23:57:21Z","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":"9b3ac1e631028645c4f2f77f7b1033aa2429571b6f257d3d903a7b2e6a28e50f","cross_cats_sorted":["cs.AI","math.OC","stat.ML"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2018-11-21T18:58:46Z","title_canon_sha256":"ba3d66e38335314c46352c85f975f0a56f2edf99fa82881beb3d59975108a5c7"},"schema_version":"1.0","source":{"id":"1811.08888","kind":"arxiv","version":3}},"canonical_sha256":"62756077e38369dd08ca5c0d48920fb08e08ea73a30ffc793368bcb2808112f5","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"62756077e38369dd08ca5c0d48920fb08e08ea73a30ffc793368bcb2808112f5","first_computed_at":"2026-05-17T23:57:21.181083Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-17T23:57:21.181083Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"9rGZulk9NOyfgjf3SYJdn2Glu6yZLbaYVcifoKVKzM2LNgYhW+l8vUUq5X5mxJaCjk+c37LB8ZXGIaauX8GcAw==","signature_status":"signed_v1","signed_at":"2026-05-17T23:57:21.181773Z","signed_message":"canonical_sha256_bytes"},"source_id":"1811.08888","source_kind":"arxiv","source_version":3}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:99796a43c450e8636dffe4f25f75c5ed3da6bb0857fc4abaa354714312e37ca3","sha256:945b8a181061d8160da1ba4a3074de1509df74a227df5835f3c7b4cbc4657976"],"state_sha256":"f1cacdb32c4ead5167c5734a8e72bbb61b5fba167ab052f9e46f22bc06a4267a"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"0gvBmtXNpKKZ9AGe3TQtLpYiFD0f0W5x6jcLE75pk/YEq8NxFvSJwq0E1q3VEBGQnlJj9b2rFhJTDh8QOo9bAA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-05-28T21:45:53.387054Z","bundle_sha256":"ae28152f48b19e798f0b55d7f955ddb1b9185db9046730f87fb5b066d059f0ea"}}