{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2019:DIMOVFKKUAVDTCQUA6EZW2EBNX","short_pith_number":"pith:DIMOVFKK","canonical_record":{"source":{"id":"1907.04595","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2019-07-10T09:47:43Z","cross_cats_sorted":["stat.ML"],"title_canon_sha256":"70b56f1fe01600fe3058258c18e3896a4f69de48a72aa705db7a76146086c6b2","abstract_canon_sha256":"9455d2aec2262deb09624253f4a5bde3aa6eba2e8ae075a90ac912e1ebea2b74"},"schema_version":"1.0"},"canonical_sha256":"1a18ea954aa02a398a1407899b68816df473cd08eae597ffb7593ae21dfe2989","source":{"kind":"arxiv","id":"1907.04595","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1907.04595","created_at":"2026-07-05T00:58:03Z"},{"alias_kind":"arxiv_version","alias_value":"1907.04595v2","created_at":"2026-07-05T00:58:03Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1907.04595","created_at":"2026-07-05T00:58:03Z"},{"alias_kind":"pith_short_12","alias_value":"DIMOVFKKUAVD","created_at":"2026-07-05T00:58:03Z"},{"alias_kind":"pith_short_16","alias_value":"DIMOVFKKUAVDTCQU","created_at":"2026-07-05T00:58:03Z"},{"alias_kind":"pith_short_8","alias_value":"DIMOVFKK","created_at":"2026-07-05T00:58:03Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2019:DIMOVFKKUAVDTCQUA6EZW2EBNX","target":"record","payload":{"canonical_record":{"source":{"id":"1907.04595","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2019-07-10T09:47:43Z","cross_cats_sorted":["stat.ML"],"title_canon_sha256":"70b56f1fe01600fe3058258c18e3896a4f69de48a72aa705db7a76146086c6b2","abstract_canon_sha256":"9455d2aec2262deb09624253f4a5bde3aa6eba2e8ae075a90ac912e1ebea2b74"},"schema_version":"1.0"},"canonical_sha256":"1a18ea954aa02a398a1407899b68816df473cd08eae597ffb7593ae21dfe2989","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T00:58:03.743443Z","signature_b64":"R2xn0z69xWJYbsZ9sbwGwuRHOiPO6viJKP7LhwCSLpaSsC6okUnJwzPbFbZThuupymlbw64x+2PIzStJuTIZBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"1a18ea954aa02a398a1407899b68816df473cd08eae597ffb7593ae21dfe2989","last_reissued_at":"2026-07-05T00:58:03.743005Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T00:58:03.743005Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"1907.04595","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":"2026-07-05T00:58:03Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"+RVXUqhAqLNqVus0G91sM6vAbC1MnhAniFIPEHEBFzn4C+ji5V+nsLxfg1iBrYGmbzbuLfWKehuC5rtGS3Q1Cw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-19T14:37:57.726891Z"},"content_sha256":"7d84099ad2d935033e8489d97b0a27a2bea2f7902466471c06090a7b29103668","schema_version":"1.0","event_id":"sha256:7d84099ad2d935033e8489d97b0a27a2bea2f7902466471c06090a7b29103668"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2019:DIMOVFKKUAVDTCQUA6EZW2EBNX","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Towards Explaining the Regularization Effect of Initial Large Learning Rate in Training Neural Networks","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Colin Wei, Tengyu Ma, Yuanzhi Li","submitted_at":"2019-07-10T09:47:43Z","abstract_excerpt":"Stochastic gradient descent with a large initial learning rate is widely used for training modern neural net architectures. Although a small initial learning rate allows for faster training and better test performance initially, the large learning rate achieves better generalization soon after the learning rate is annealed. Towards explaining this phenomenon, we devise a setting in which we can prove that a two layer network trained with large initial learning rate and annealing provably generalizes better than the same network trained with a small learning rate from the start. The key insight"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1907.04595","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/1907.04595/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":"2026-07-05T00:58:03Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"aAJ9Nid+u43aI18wYhJuCwc9NzxorElvbYl4qTQjDBYJ0H8Md6no4clPapufvJHAfmjTH+eA+kOIbz+Afw+HDg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-19T14:37:57.727560Z"},"content_sha256":"052717a0a3448b74c1742b93d8cda59093b33f4c1eb2c212c531694cb9e81311","schema_version":"1.0","event_id":"sha256:052717a0a3448b74c1742b93d8cda59093b33f4c1eb2c212c531694cb9e81311"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/DIMOVFKKUAVDTCQUA6EZW2EBNX/bundle.json","state_url":"https://pith.science/pith/DIMOVFKKUAVDTCQUA6EZW2EBNX/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/DIMOVFKKUAVDTCQUA6EZW2EBNX/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":"2026-08-19T14:37:57Z","links":{"resolver":"https://pith.science/pith/DIMOVFKKUAVDTCQUA6EZW2EBNX","bundle":"https://pith.science/pith/DIMOVFKKUAVDTCQUA6EZW2EBNX/bundle.json","state":"https://pith.science/pith/DIMOVFKKUAVDTCQUA6EZW2EBNX/state.json","well_known_bundle":"https://pith.science/.well-known/pith/DIMOVFKKUAVDTCQUA6EZW2EBNX/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2019:DIMOVFKKUAVDTCQUA6EZW2EBNX","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":"9455d2aec2262deb09624253f4a5bde3aa6eba2e8ae075a90ac912e1ebea2b74","cross_cats_sorted":["stat.ML"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2019-07-10T09:47:43Z","title_canon_sha256":"70b56f1fe01600fe3058258c18e3896a4f69de48a72aa705db7a76146086c6b2"},"schema_version":"1.0","source":{"id":"1907.04595","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1907.04595","created_at":"2026-07-05T00:58:03Z"},{"alias_kind":"arxiv_version","alias_value":"1907.04595v2","created_at":"2026-07-05T00:58:03Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1907.04595","created_at":"2026-07-05T00:58:03Z"},{"alias_kind":"pith_short_12","alias_value":"DIMOVFKKUAVD","created_at":"2026-07-05T00:58:03Z"},{"alias_kind":"pith_short_16","alias_value":"DIMOVFKKUAVDTCQU","created_at":"2026-07-05T00:58:03Z"},{"alias_kind":"pith_short_8","alias_value":"DIMOVFKK","created_at":"2026-07-05T00:58:03Z"}],"graph_snapshots":[{"event_id":"sha256:052717a0a3448b74c1742b93d8cda59093b33f4c1eb2c212c531694cb9e81311","target":"graph","created_at":"2026-07-05T00:58:03Z","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/1907.04595/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Stochastic gradient descent with a large initial learning rate is widely used for training modern neural net architectures. Although a small initial learning rate allows for faster training and better test performance initially, the large learning rate achieves better generalization soon after the learning rate is annealed. Towards explaining this phenomenon, we devise a setting in which we can prove that a two layer network trained with large initial learning rate and annealing provably generalizes better than the same network trained with a small learning rate from the start. The key insight","authors_text":"Colin Wei, Tengyu Ma, Yuanzhi Li","cross_cats":["stat.ML"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2019-07-10T09:47:43Z","title":"Towards Explaining the Regularization Effect of Initial Large Learning Rate in Training Neural Networks"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1907.04595","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":"sha256:7d84099ad2d935033e8489d97b0a27a2bea2f7902466471c06090a7b29103668","target":"record","created_at":"2026-07-05T00:58:03Z","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":"9455d2aec2262deb09624253f4a5bde3aa6eba2e8ae075a90ac912e1ebea2b74","cross_cats_sorted":["stat.ML"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2019-07-10T09:47:43Z","title_canon_sha256":"70b56f1fe01600fe3058258c18e3896a4f69de48a72aa705db7a76146086c6b2"},"schema_version":"1.0","source":{"id":"1907.04595","kind":"arxiv","version":2}},"canonical_sha256":"1a18ea954aa02a398a1407899b68816df473cd08eae597ffb7593ae21dfe2989","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"1a18ea954aa02a398a1407899b68816df473cd08eae597ffb7593ae21dfe2989","first_computed_at":"2026-07-05T00:58:03.743005Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T00:58:03.743005Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"R2xn0z69xWJYbsZ9sbwGwuRHOiPO6viJKP7LhwCSLpaSsC6okUnJwzPbFbZThuupymlbw64x+2PIzStJuTIZBg==","signature_status":"signed_v1","signed_at":"2026-07-05T00:58:03.743443Z","signed_message":"canonical_sha256_bytes"},"source_id":"1907.04595","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:7d84099ad2d935033e8489d97b0a27a2bea2f7902466471c06090a7b29103668","sha256:052717a0a3448b74c1742b93d8cda59093b33f4c1eb2c212c531694cb9e81311"],"state_sha256":"a2fccabc3ad26fcf1bd173f4b2a6d29e1b1a5ef5e259a8c0bd8e4bbc5fbeb3e7"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"sxKQmDDBRlvjI9FzX4V1sOqzP6Pexfo0W5CuUyELOCjqSki20oumwBbOON3iu2eCFsW+c8v9B+W2XrAXQX+zCQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-19T14:37:57.733730Z","bundle_sha256":"efaf2fe2541eae8f27fdc66eee290eff77bc3188cb95b073ace7410b4a52cbeb"}}