{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2020:W33LUAQO7EHS47CN2IMOHX4RCB","short_pith_number":"pith:W33LUAQO","canonical_record":{"source":{"id":"2007.04596","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2020-07-09T07:09:28Z","cross_cats_sorted":["math.OC","stat.ML"],"title_canon_sha256":"4c2eb6b37cdaa316c8e2dfaaee5875f159ef870b4ba85deb37d65d821bff4575","abstract_canon_sha256":"36c7f9632555b4ae4101fd9cc8e7e7fad82c99fb739318010c0f62edbed24955"},"schema_version":"1.0"},"canonical_sha256":"b6f6ba020ef90f2e7c4dd218e3df911047a9b3a63f223181b784a9ff1f07ef06","source":{"kind":"arxiv","id":"2007.04596","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2007.04596","created_at":"2026-07-05T01:17:31Z"},{"alias_kind":"arxiv_version","alias_value":"2007.04596v1","created_at":"2026-07-05T01:17:31Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2007.04596","created_at":"2026-07-05T01:17:31Z"},{"alias_kind":"pith_short_12","alias_value":"W33LUAQO7EHS","created_at":"2026-07-05T01:17:31Z"},{"alias_kind":"pith_short_16","alias_value":"W33LUAQO7EHS47CN","created_at":"2026-07-05T01:17:31Z"},{"alias_kind":"pith_short_8","alias_value":"W33LUAQO","created_at":"2026-07-05T01:17:31Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2020:W33LUAQO7EHS47CN2IMOHX4RCB","target":"record","payload":{"canonical_record":{"source":{"id":"2007.04596","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2020-07-09T07:09:28Z","cross_cats_sorted":["math.OC","stat.ML"],"title_canon_sha256":"4c2eb6b37cdaa316c8e2dfaaee5875f159ef870b4ba85deb37d65d821bff4575","abstract_canon_sha256":"36c7f9632555b4ae4101fd9cc8e7e7fad82c99fb739318010c0f62edbed24955"},"schema_version":"1.0"},"canonical_sha256":"b6f6ba020ef90f2e7c4dd218e3df911047a9b3a63f223181b784a9ff1f07ef06","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T01:17:31.563306Z","signature_b64":"qr3bbJjMjAo5Zn1hYsYCze1QDn5yivmbsd0tDOGDzSYrqeQpMuT2d47oq2jl2tOL9G236LXfOyqqoHu3uDunAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"b6f6ba020ef90f2e7c4dd218e3df911047a9b3a63f223181b784a9ff1f07ef06","last_reissued_at":"2026-07-05T01:17:31.562907Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T01:17:31.562907Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2007.04596","source_version":1,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T01:17:31Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"e+THKThOtRML/ve+tTJwUCbYu9EjDLHEB9o0M2FeO5P4RdRRIGGZoP8LJDbbgBBuPHo/PP9N29H+Aba8UBEFBA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-17T18:14:57.150988Z"},"content_sha256":"0a713a48312fe566a081d6022fa08c46683d1e225d38afe63b5ac0dd8aff5343","schema_version":"1.0","event_id":"sha256:0a713a48312fe566a081d6022fa08c46683d1e225d38afe63b5ac0dd8aff5343"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2020:W33LUAQO7EHS47CN2IMOHX4RCB","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Learning Over-Parametrized Two-Layer ReLU Neural Networks beyond NTK","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["math.OC","stat.ML"],"primary_cat":"cs.LG","authors_text":"Hongyang R. Zhang, Tengyu Ma, Yuanzhi Li","submitted_at":"2020-07-09T07:09:28Z","abstract_excerpt":"We consider the dynamic of gradient descent for learning a two-layer neural network. We assume the input $x\\in\\mathbb{R}^d$ is drawn from a Gaussian distribution and the label of $x$ satisfies $f^{\\star}(x) = a^{\\top}|W^{\\star}x|$, where $a\\in\\mathbb{R}^d$ is a nonnegative vector and $W^{\\star} \\in\\mathbb{R}^{d\\times d}$ is an orthonormal matrix. We show that an over-parametrized two-layer neural network with ReLU activation, trained by gradient descent from random initialization, can provably learn the ground truth network with population loss at most $o(1/d)$ in polynomial time with polynomi"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2007.04596","kind":"arxiv","version":1},"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/2007.04596/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":"2026-07-05T01:17:31Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"TK6sn6kWvEI08XlpFNN0wXkXXmMtya7L18w7TxF5xL3Myz97E6XFaOWsX8VBwPtfoV9Ey2/1jH4BjQeJaz2zAA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-17T18:14:57.151984Z"},"content_sha256":"e9028f4221c4cffe3050524c2654ccf04d6cca15e7cbcdce9bca36a4fa4b243c","schema_version":"1.0","event_id":"sha256:e9028f4221c4cffe3050524c2654ccf04d6cca15e7cbcdce9bca36a4fa4b243c"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/W33LUAQO7EHS47CN2IMOHX4RCB/bundle.json","state_url":"https://pith.science/pith/W33LUAQO7EHS47CN2IMOHX4RCB/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/W33LUAQO7EHS47CN2IMOHX4RCB/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":"2026-08-17T18:14:57Z","links":{"resolver":"https://pith.science/pith/W33LUAQO7EHS47CN2IMOHX4RCB","bundle":"https://pith.science/pith/W33LUAQO7EHS47CN2IMOHX4RCB/bundle.json","state":"https://pith.science/pith/W33LUAQO7EHS47CN2IMOHX4RCB/state.json","well_known_bundle":"https://pith.science/.well-known/pith/W33LUAQO7EHS47CN2IMOHX4RCB/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2020:W33LUAQO7EHS47CN2IMOHX4RCB","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":"36c7f9632555b4ae4101fd9cc8e7e7fad82c99fb739318010c0f62edbed24955","cross_cats_sorted":["math.OC","stat.ML"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2020-07-09T07:09:28Z","title_canon_sha256":"4c2eb6b37cdaa316c8e2dfaaee5875f159ef870b4ba85deb37d65d821bff4575"},"schema_version":"1.0","source":{"id":"2007.04596","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2007.04596","created_at":"2026-07-05T01:17:31Z"},{"alias_kind":"arxiv_version","alias_value":"2007.04596v1","created_at":"2026-07-05T01:17:31Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2007.04596","created_at":"2026-07-05T01:17:31Z"},{"alias_kind":"pith_short_12","alias_value":"W33LUAQO7EHS","created_at":"2026-07-05T01:17:31Z"},{"alias_kind":"pith_short_16","alias_value":"W33LUAQO7EHS47CN","created_at":"2026-07-05T01:17:31Z"},{"alias_kind":"pith_short_8","alias_value":"W33LUAQO","created_at":"2026-07-05T01:17:31Z"}],"graph_snapshots":[{"event_id":"sha256:e9028f4221c4cffe3050524c2654ccf04d6cca15e7cbcdce9bca36a4fa4b243c","target":"graph","created_at":"2026-07-05T01:17:31Z","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/2007.04596/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"We consider the dynamic of gradient descent for learning a two-layer neural network. We assume the input $x\\in\\mathbb{R}^d$ is drawn from a Gaussian distribution and the label of $x$ satisfies $f^{\\star}(x) = a^{\\top}|W^{\\star}x|$, where $a\\in\\mathbb{R}^d$ is a nonnegative vector and $W^{\\star} \\in\\mathbb{R}^{d\\times d}$ is an orthonormal matrix. We show that an over-parametrized two-layer neural network with ReLU activation, trained by gradient descent from random initialization, can provably learn the ground truth network with population loss at most $o(1/d)$ in polynomial time with polynomi","authors_text":"Hongyang R. Zhang, Tengyu Ma, Yuanzhi Li","cross_cats":["math.OC","stat.ML"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2020-07-09T07:09:28Z","title":"Learning Over-Parametrized Two-Layer ReLU Neural Networks beyond NTK"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2007.04596","kind":"arxiv","version":1},"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":"sha256:0a713a48312fe566a081d6022fa08c46683d1e225d38afe63b5ac0dd8aff5343","target":"record","created_at":"2026-07-05T01:17:31Z","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":"36c7f9632555b4ae4101fd9cc8e7e7fad82c99fb739318010c0f62edbed24955","cross_cats_sorted":["math.OC","stat.ML"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2020-07-09T07:09:28Z","title_canon_sha256":"4c2eb6b37cdaa316c8e2dfaaee5875f159ef870b4ba85deb37d65d821bff4575"},"schema_version":"1.0","source":{"id":"2007.04596","kind":"arxiv","version":1}},"canonical_sha256":"b6f6ba020ef90f2e7c4dd218e3df911047a9b3a63f223181b784a9ff1f07ef06","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"b6f6ba020ef90f2e7c4dd218e3df911047a9b3a63f223181b784a9ff1f07ef06","first_computed_at":"2026-07-05T01:17:31.562907Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T01:17:31.562907Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"qr3bbJjMjAo5Zn1hYsYCze1QDn5yivmbsd0tDOGDzSYrqeQpMuT2d47oq2jl2tOL9G236LXfOyqqoHu3uDunAQ==","signature_status":"signed_v1","signed_at":"2026-07-05T01:17:31.563306Z","signed_message":"canonical_sha256_bytes"},"source_id":"2007.04596","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:0a713a48312fe566a081d6022fa08c46683d1e225d38afe63b5ac0dd8aff5343","sha256:e9028f4221c4cffe3050524c2654ccf04d6cca15e7cbcdce9bca36a4fa4b243c"],"state_sha256":"1e463aec6bb3e8c1dc5814c8612fa94dc47111a120f6a0e27ef0aecd071b1150"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"6QIG5JrWfy3og4leoOuaTjDKkH3qQetKOf4GMEVRnms6n0KYMmVcjhHN1QWrTMVWxFRbJdtj2w7+dOc/By+VBw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-17T18:14:57.160933Z","bundle_sha256":"090fb47a435a9ecf215c6daca68ac9e2b55edfe5c7cdab2dca0c364ed9f066f4"}}