{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:HW5BVY32KABAPMMKRO4PXCKMSM","short_pith_number":"pith:HW5BVY32","schema_version":"1.0","canonical_sha256":"3dba1ae37a500207b18a8bb8fb894c932be11a1bfb7871497d025e3df7673133","source":{"kind":"arxiv","id":"2411.16914","version":1},"attestation_state":"computed","paper":{"title":"Curvature in the Looking-Glass: Optimal Methods to Exploit Curvature of Expectation in the Loss Landscape","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Alexander Safonov, Jed A. Duersch, Jeremy Wendt, Tommie A. Catanach","submitted_at":"2024-11-25T20:32:57Z","abstract_excerpt":"Harnessing the local topography of the loss landscape is a central challenge in advanced optimization tasks. By accounting for the effect of potential parameter changes, we can alter the model more efficiently. Contrary to standard assumptions, we find that the Hessian does not always approximate loss curvature well, particularly near gradient discontinuities, which commonly arise in deep learning architectures.\n  We present a new conceptual framework to understand how curvature of expected changes in loss emerges in architectures with many rectified linear units. Each ReLU creates a parameter"},"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":"2411.16914","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2024-11-25T20:32:57Z","cross_cats_sorted":["stat.ML"],"title_canon_sha256":"2de672375e9faffa573b9ca7c406fc706fe783d981b812739158955e47f97bd5","abstract_canon_sha256":"b56eabfefdbe62883f5d8ed2c798abefa775053cbc70a956c1d53e7929d93d53"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:40:24.553208Z","signature_b64":"Mirdw2PzCDhYoyz0UesfLXw7SUUxvyQi62cVGgwFMhV1N0glINOGu2hk3qvBdBB8f7xX67PXf1GbXt7Q9mKaDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"3dba1ae37a500207b18a8bb8fb894c932be11a1bfb7871497d025e3df7673133","last_reissued_at":"2026-07-05T09:40:24.552748Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:40:24.552748Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Curvature in the Looking-Glass: Optimal Methods to Exploit Curvature of Expectation in the Loss Landscape","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Alexander Safonov, Jed A. Duersch, Jeremy Wendt, Tommie A. Catanach","submitted_at":"2024-11-25T20:32:57Z","abstract_excerpt":"Harnessing the local topography of the loss landscape is a central challenge in advanced optimization tasks. By accounting for the effect of potential parameter changes, we can alter the model more efficiently. Contrary to standard assumptions, we find that the Hessian does not always approximate loss curvature well, particularly near gradient discontinuities, which commonly arise in deep learning architectures.\n  We present a new conceptual framework to understand how curvature of expected changes in loss emerges in architectures with many rectified linear units. Each ReLU creates a parameter"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2411.16914","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/2411.16914/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":"2411.16914","created_at":"2026-07-05T09:40:24.552803+00:00"},{"alias_kind":"arxiv_version","alias_value":"2411.16914v1","created_at":"2026-07-05T09:40:24.552803+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2411.16914","created_at":"2026-07-05T09:40:24.552803+00:00"},{"alias_kind":"pith_short_12","alias_value":"HW5BVY32KABA","created_at":"2026-07-05T09:40:24.552803+00:00"},{"alias_kind":"pith_short_16","alias_value":"HW5BVY32KABAPMMK","created_at":"2026-07-05T09:40:24.552803+00:00"},{"alias_kind":"pith_short_8","alias_value":"HW5BVY32","created_at":"2026-07-05T09:40:24.552803+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/HW5BVY32KABAPMMKRO4PXCKMSM","json":"https://pith.science/pith/HW5BVY32KABAPMMKRO4PXCKMSM.json","graph_json":"https://pith.science/api/pith-number/HW5BVY32KABAPMMKRO4PXCKMSM/graph.json","events_json":"https://pith.science/api/pith-number/HW5BVY32KABAPMMKRO4PXCKMSM/events.json","paper":"https://pith.science/paper/HW5BVY32"},"agent_actions":{"view_html":"https://pith.science/pith/HW5BVY32KABAPMMKRO4PXCKMSM","download_json":"https://pith.science/pith/HW5BVY32KABAPMMKRO4PXCKMSM.json","view_paper":"https://pith.science/paper/HW5BVY32","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2411.16914&json=true","fetch_graph":"https://pith.science/api/pith-number/HW5BVY32KABAPMMKRO4PXCKMSM/graph.json","fetch_events":"https://pith.science/api/pith-number/HW5BVY32KABAPMMKRO4PXCKMSM/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/HW5BVY32KABAPMMKRO4PXCKMSM/action/timestamp_anchor","attest_storage":"https://pith.science/pith/HW5BVY32KABAPMMKRO4PXCKMSM/action/storage_attestation","attest_author":"https://pith.science/pith/HW5BVY32KABAPMMKRO4PXCKMSM/action/author_attestation","sign_citation":"https://pith.science/pith/HW5BVY32KABAPMMKRO4PXCKMSM/action/citation_signature","submit_replication":"https://pith.science/pith/HW5BVY32KABAPMMKRO4PXCKMSM/action/replication_record"}},"created_at":"2026-07-05T09:40:24.552803+00:00","updated_at":"2026-07-05T09:40:24.552803+00:00"}