{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2019:NXQE3RFCH4RRMJCDVYH5O6THSX","short_pith_number":"pith:NXQE3RFC","canonical_record":{"source":{"id":"1901.08278","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2019-01-24T08:27:03Z","cross_cats_sorted":["cs.CV","stat.ML"],"title_canon_sha256":"01fe3992b180d9912c75b9ecc4ced60ff629538b639ac3f389aee710443d52f3","abstract_canon_sha256":"8802196f65059412df92fbc773877257c8d5b3e13d1fd22835d5117e6b6cb032"},"schema_version":"1.0"},"canonical_sha256":"6de04dc4a23f23162443ae0fd77a6795cd291f2a31b2a6814fdaec6caf30ab8f","source":{"kind":"arxiv","id":"1901.08278","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1901.08278","created_at":"2026-07-05T00:35:42Z"},{"alias_kind":"arxiv_version","alias_value":"1901.08278v2","created_at":"2026-07-05T00:35:42Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1901.08278","created_at":"2026-07-05T00:35:42Z"},{"alias_kind":"pith_short_12","alias_value":"NXQE3RFCH4RR","created_at":"2026-07-05T00:35:42Z"},{"alias_kind":"pith_short_16","alias_value":"NXQE3RFCH4RRMJCD","created_at":"2026-07-05T00:35:42Z"},{"alias_kind":"pith_short_8","alias_value":"NXQE3RFC","created_at":"2026-07-05T00:35:42Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2019:NXQE3RFCH4RRMJCDVYH5O6THSX","target":"record","payload":{"canonical_record":{"source":{"id":"1901.08278","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2019-01-24T08:27:03Z","cross_cats_sorted":["cs.CV","stat.ML"],"title_canon_sha256":"01fe3992b180d9912c75b9ecc4ced60ff629538b639ac3f389aee710443d52f3","abstract_canon_sha256":"8802196f65059412df92fbc773877257c8d5b3e13d1fd22835d5117e6b6cb032"},"schema_version":"1.0"},"canonical_sha256":"6de04dc4a23f23162443ae0fd77a6795cd291f2a31b2a6814fdaec6caf30ab8f","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T00:35:42.733566Z","signature_b64":"1slNB+9JhgHF6qe+2jr/yv6nX2O4utBUpml8WWnIjScSHuNwU3cUxAldBnWCKsi5RZs13/tRbSWlxdgQqNp4CQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"6de04dc4a23f23162443ae0fd77a6795cd291f2a31b2a6814fdaec6caf30ab8f","last_reissued_at":"2026-07-05T00:35:42.733129Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T00:35:42.733129Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"1901.08278","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":"2026-07-05T00:35:42Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"f4xZhsyiP1U6S5TAlbS5QVmrMsHrYVuHkdDNHN7UMgeqb2gvFdbaCGv90ePHsENRU2N5fUxC+NnUHR2Mh6ibAg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-08T21:48:56.401508Z"},"content_sha256":"80b3734102fd0a8a22d1bc4169a8e0b2d847bd7b4d278f7b67375f882b9867e5","schema_version":"1.0","event_id":"sha256:80b3734102fd0a8a22d1bc4169a8e0b2d847bd7b4d278f7b67375f882b9867e5"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2019:NXQE3RFCH4RRMJCDVYH5O6THSX","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Heavy-Tailed Universality Predicts Trends in Test Accuracies for Very Large Pre-Trained Deep Neural Networks","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CV","stat.ML"],"primary_cat":"cs.LG","authors_text":"Charles H. Martin, Michael W. Mahoney","submitted_at":"2019-01-24T08:27:03Z","abstract_excerpt":"Given two or more Deep Neural Networks (DNNs) with the same or similar architectures, and trained on the same dataset, but trained with different solvers, parameters, hyper-parameters, regularization, etc., can we predict which DNN will have the best test accuracy, and can we do so without peeking at the test data? In this paper, we show how to use a new Theory of Heavy-Tailed Self-Regularization (HT-SR) to answer this. HT-SR suggests, among other things, that modern DNNs exhibit what we call Heavy-Tailed Mechanistic Universality (HT-MU), meaning that the correlations in the layer weight matri"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1901.08278","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/1901.08278/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":"2026-07-05T00:35:42Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"GQkbUoLFJGHEnbBXNNJwDiJwdchKqxR3tUI5lRISsx9mkdfm5PsfBLawvyzhY8fC4ZSjkhjQ/+tg5KYpS9sPAg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-08T21:48:56.402115Z"},"content_sha256":"75f8f34eb9731436b0f6f4c2109ca133fa2f683900561e4ad36b809ab519c31e","schema_version":"1.0","event_id":"sha256:75f8f34eb9731436b0f6f4c2109ca133fa2f683900561e4ad36b809ab519c31e"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/NXQE3RFCH4RRMJCDVYH5O6THSX/bundle.json","state_url":"https://pith.science/pith/NXQE3RFCH4RRMJCDVYH5O6THSX/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/NXQE3RFCH4RRMJCDVYH5O6THSX/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":"2026-08-08T21:48:56Z","links":{"resolver":"https://pith.science/pith/NXQE3RFCH4RRMJCDVYH5O6THSX","bundle":"https://pith.science/pith/NXQE3RFCH4RRMJCDVYH5O6THSX/bundle.json","state":"https://pith.science/pith/NXQE3RFCH4RRMJCDVYH5O6THSX/state.json","well_known_bundle":"https://pith.science/.well-known/pith/NXQE3RFCH4RRMJCDVYH5O6THSX/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2019:NXQE3RFCH4RRMJCDVYH5O6THSX","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":"8802196f65059412df92fbc773877257c8d5b3e13d1fd22835d5117e6b6cb032","cross_cats_sorted":["cs.CV","stat.ML"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2019-01-24T08:27:03Z","title_canon_sha256":"01fe3992b180d9912c75b9ecc4ced60ff629538b639ac3f389aee710443d52f3"},"schema_version":"1.0","source":{"id":"1901.08278","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1901.08278","created_at":"2026-07-05T00:35:42Z"},{"alias_kind":"arxiv_version","alias_value":"1901.08278v2","created_at":"2026-07-05T00:35:42Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1901.08278","created_at":"2026-07-05T00:35:42Z"},{"alias_kind":"pith_short_12","alias_value":"NXQE3RFCH4RR","created_at":"2026-07-05T00:35:42Z"},{"alias_kind":"pith_short_16","alias_value":"NXQE3RFCH4RRMJCD","created_at":"2026-07-05T00:35:42Z"},{"alias_kind":"pith_short_8","alias_value":"NXQE3RFC","created_at":"2026-07-05T00:35:42Z"}],"graph_snapshots":[{"event_id":"sha256:75f8f34eb9731436b0f6f4c2109ca133fa2f683900561e4ad36b809ab519c31e","target":"graph","created_at":"2026-07-05T00:35:42Z","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/1901.08278/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Given two or more Deep Neural Networks (DNNs) with the same or similar architectures, and trained on the same dataset, but trained with different solvers, parameters, hyper-parameters, regularization, etc., can we predict which DNN will have the best test accuracy, and can we do so without peeking at the test data? In this paper, we show how to use a new Theory of Heavy-Tailed Self-Regularization (HT-SR) to answer this. HT-SR suggests, among other things, that modern DNNs exhibit what we call Heavy-Tailed Mechanistic Universality (HT-MU), meaning that the correlations in the layer weight matri","authors_text":"Charles H. Martin, Michael W. Mahoney","cross_cats":["cs.CV","stat.ML"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2019-01-24T08:27:03Z","title":"Heavy-Tailed Universality Predicts Trends in Test Accuracies for Very Large Pre-Trained Deep Neural Networks"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1901.08278","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":"sha256:80b3734102fd0a8a22d1bc4169a8e0b2d847bd7b4d278f7b67375f882b9867e5","target":"record","created_at":"2026-07-05T00:35:42Z","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":"8802196f65059412df92fbc773877257c8d5b3e13d1fd22835d5117e6b6cb032","cross_cats_sorted":["cs.CV","stat.ML"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2019-01-24T08:27:03Z","title_canon_sha256":"01fe3992b180d9912c75b9ecc4ced60ff629538b639ac3f389aee710443d52f3"},"schema_version":"1.0","source":{"id":"1901.08278","kind":"arxiv","version":2}},"canonical_sha256":"6de04dc4a23f23162443ae0fd77a6795cd291f2a31b2a6814fdaec6caf30ab8f","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"6de04dc4a23f23162443ae0fd77a6795cd291f2a31b2a6814fdaec6caf30ab8f","first_computed_at":"2026-07-05T00:35:42.733129Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T00:35:42.733129Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"1slNB+9JhgHF6qe+2jr/yv6nX2O4utBUpml8WWnIjScSHuNwU3cUxAldBnWCKsi5RZs13/tRbSWlxdgQqNp4CQ==","signature_status":"signed_v1","signed_at":"2026-07-05T00:35:42.733566Z","signed_message":"canonical_sha256_bytes"},"source_id":"1901.08278","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:80b3734102fd0a8a22d1bc4169a8e0b2d847bd7b4d278f7b67375f882b9867e5","sha256:75f8f34eb9731436b0f6f4c2109ca133fa2f683900561e4ad36b809ab519c31e"],"state_sha256":"e27b18fc06c94f3d9f570f07c43e80886adbbb012b6734958a0e08aeb9f3eed8"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"yN4RRjNot27yGKpuTzHYbcLL8hUr2AMq5bwUEIv/n2YYZqWO+H08ORGbDM2OPt8n8vXTsH4ziVmExiUt6B5GCA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-08T21:48:56.406165Z","bundle_sha256":"d690b84b3662cba1ce103e39d96e6e706fc6b2a1e8ce26a0b7be9ccc4c91a070"}}