{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:LTAZE7WI537OE77YTWR3PVPS47","short_pith_number":"pith:LTAZE7WI","schema_version":"1.0","canonical_sha256":"5cc1927ec8eefee27ff89da3b7d5f2e7cd516728aca8c928c2b9262f37bd2089","source":{"kind":"arxiv","id":"2305.14683","version":4},"attestation_state":"computed","paper":{"title":"On progressive sharpening, flat minima and generalisation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Jack Valmadre, Lachlan Ewen MacDonald, Simon Lucey","submitted_at":"2023-05-24T03:44:50Z","abstract_excerpt":"We present a new approach to understanding the relationship between loss curvature and input-output model behaviour in deep learning. Specifically, we use existing empirical analyses of the spectrum of deep network loss Hessians to ground an ansatz tying together the loss Hessian and the input-output Jacobian over training samples during the training of deep neural networks. We then prove a series of theoretical results which quantify the degree to which the input-output Jacobian of a model approximates its Lipschitz norm over a data distribution, and deduce a novel generalisation bound in ter"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2305.14683","kind":"arxiv","version":4},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2023-05-24T03:44:50Z","cross_cats_sorted":["stat.ML"],"title_canon_sha256":"63400a0daa7f714fa3ff07ad593572d20e50edfbc39271f8fc23ca0b24f3d2ca","abstract_canon_sha256":"21a5c97fbe21633c425321b274cf894f0100d49d2479e7fcbe01ccccfc018e26"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T06:54:38.878256Z","signature_b64":"fT1L2XCHjqyECPHFGX5UYdgjH/S4KItaSX4wN7J96u/ZAYLXJmVixiLs+TrdeY1PUZy9GH7GriUo1uY+TiTMAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"5cc1927ec8eefee27ff89da3b7d5f2e7cd516728aca8c928c2b9262f37bd2089","last_reissued_at":"2026-07-05T06:54:38.877812Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T06:54:38.877812Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"On progressive sharpening, flat minima and generalisation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Jack Valmadre, Lachlan Ewen MacDonald, Simon Lucey","submitted_at":"2023-05-24T03:44:50Z","abstract_excerpt":"We present a new approach to understanding the relationship between loss curvature and input-output model behaviour in deep learning. Specifically, we use existing empirical analyses of the spectrum of deep network loss Hessians to ground an ansatz tying together the loss Hessian and the input-output Jacobian over training samples during the training of deep neural networks. We then prove a series of theoretical results which quantify the degree to which the input-output Jacobian of a model approximates its Lipschitz norm over a data distribution, and deduce a novel generalisation bound in ter"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2305.14683","kind":"arxiv","version":4},"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/2305.14683/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"},"aliases":[{"alias_kind":"arxiv","alias_value":"2305.14683","created_at":"2026-07-05T06:54:38.877869+00:00"},{"alias_kind":"arxiv_version","alias_value":"2305.14683v4","created_at":"2026-07-05T06:54:38.877869+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2305.14683","created_at":"2026-07-05T06:54:38.877869+00:00"},{"alias_kind":"pith_short_12","alias_value":"LTAZE7WI537O","created_at":"2026-07-05T06:54:38.877869+00:00"},{"alias_kind":"pith_short_16","alias_value":"LTAZE7WI537OE77Y","created_at":"2026-07-05T06:54:38.877869+00:00"},{"alias_kind":"pith_short_8","alias_value":"LTAZE7WI","created_at":"2026-07-05T06:54:38.877869+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/LTAZE7WI537OE77YTWR3PVPS47","json":"https://pith.science/pith/LTAZE7WI537OE77YTWR3PVPS47.json","graph_json":"https://pith.science/api/pith-number/LTAZE7WI537OE77YTWR3PVPS47/graph.json","events_json":"https://pith.science/api/pith-number/LTAZE7WI537OE77YTWR3PVPS47/events.json","paper":"https://pith.science/paper/LTAZE7WI"},"agent_actions":{"view_html":"https://pith.science/pith/LTAZE7WI537OE77YTWR3PVPS47","download_json":"https://pith.science/pith/LTAZE7WI537OE77YTWR3PVPS47.json","view_paper":"https://pith.science/paper/LTAZE7WI","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2305.14683&json=true","fetch_graph":"https://pith.science/api/pith-number/LTAZE7WI537OE77YTWR3PVPS47/graph.json","fetch_events":"https://pith.science/api/pith-number/LTAZE7WI537OE77YTWR3PVPS47/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/LTAZE7WI537OE77YTWR3PVPS47/action/timestamp_anchor","attest_storage":"https://pith.science/pith/LTAZE7WI537OE77YTWR3PVPS47/action/storage_attestation","attest_author":"https://pith.science/pith/LTAZE7WI537OE77YTWR3PVPS47/action/author_attestation","sign_citation":"https://pith.science/pith/LTAZE7WI537OE77YTWR3PVPS47/action/citation_signature","submit_replication":"https://pith.science/pith/LTAZE7WI537OE77YTWR3PVPS47/action/replication_record"}},"created_at":"2026-07-05T06:54:38.877869+00:00","updated_at":"2026-07-05T06:54:38.877869+00:00"}