{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:A7I7HD76JQAMWBIJX4MILLBYAG","short_pith_number":"pith:A7I7HD76","schema_version":"1.0","canonical_sha256":"07d1f38ffe4c00cb0509bf1885ac3801a4d3be67489ce8918287b2fb9fccaa43","source":{"kind":"arxiv","id":"2404.03290","version":2},"attestation_state":"computed","paper":{"title":"Learning-to-Optimize with PAC-Bayesian Guarantees: Theoretical Considerations and Practical Implementation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["math.OC"],"primary_cat":"cs.LG","authors_text":"Jalal Fadili, Michael Sucker, Peter Ochs","submitted_at":"2024-04-04T08:24:57Z","abstract_excerpt":"We use the PAC-Bayesian theory for the setting of learning-to-optimize. To the best of our knowledge, we present the first framework to learn optimization algorithms with provable generalization guarantees (PAC-Bayesian bounds) and explicit trade-off between convergence guarantees and convergence speed, which contrasts with the typical worst-case analysis. Our learned optimization algorithms provably outperform related ones derived from a (deterministic) worst-case analysis. The results rely on PAC-Bayesian bounds for general, possibly unbounded loss-functions based on exponential families. Th"},"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":"2404.03290","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-04-04T08:24:57Z","cross_cats_sorted":["math.OC"],"title_canon_sha256":"a484187555e62ec8d1eb1811e036be266712abfcd5956f1df10a80844eaeaaea","abstract_canon_sha256":"c0cab153c3dff88b9875c819421e5010604daf93fae8edcdabf2c5210ed5ff0b"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:19:23.186913Z","signature_b64":"mHUa8UoB8KxAfDlnDiX6UANUvrLuNAA7UGaTg3+H0SrQtRK2lO5xU3wJ9XuA6fzES8lse9DSaas63/K1meCJBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"07d1f38ffe4c00cb0509bf1885ac3801a4d3be67489ce8918287b2fb9fccaa43","last_reissued_at":"2026-07-05T10:19:23.186403Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:19:23.186403Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Learning-to-Optimize with PAC-Bayesian Guarantees: Theoretical Considerations and Practical Implementation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["math.OC"],"primary_cat":"cs.LG","authors_text":"Jalal Fadili, Michael Sucker, Peter Ochs","submitted_at":"2024-04-04T08:24:57Z","abstract_excerpt":"We use the PAC-Bayesian theory for the setting of learning-to-optimize. To the best of our knowledge, we present the first framework to learn optimization algorithms with provable generalization guarantees (PAC-Bayesian bounds) and explicit trade-off between convergence guarantees and convergence speed, which contrasts with the typical worst-case analysis. Our learned optimization algorithms provably outperform related ones derived from a (deterministic) worst-case analysis. The results rely on PAC-Bayesian bounds for general, possibly unbounded loss-functions based on exponential families. Th"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2404.03290","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/2404.03290/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":"2404.03290","created_at":"2026-07-05T10:19:23.186471+00:00"},{"alias_kind":"arxiv_version","alias_value":"2404.03290v2","created_at":"2026-07-05T10:19:23.186471+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2404.03290","created_at":"2026-07-05T10:19:23.186471+00:00"},{"alias_kind":"pith_short_12","alias_value":"A7I7HD76JQAM","created_at":"2026-07-05T10:19:23.186471+00:00"},{"alias_kind":"pith_short_16","alias_value":"A7I7HD76JQAMWBIJ","created_at":"2026-07-05T10:19:23.186471+00:00"},{"alias_kind":"pith_short_8","alias_value":"A7I7HD76","created_at":"2026-07-05T10:19:23.186471+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2412.12156","citing_title":"Deep Distributed Optimization for Large-Scale Quadratic Programming","ref_index":53,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/A7I7HD76JQAMWBIJX4MILLBYAG","json":"https://pith.science/pith/A7I7HD76JQAMWBIJX4MILLBYAG.json","graph_json":"https://pith.science/api/pith-number/A7I7HD76JQAMWBIJX4MILLBYAG/graph.json","events_json":"https://pith.science/api/pith-number/A7I7HD76JQAMWBIJX4MILLBYAG/events.json","paper":"https://pith.science/paper/A7I7HD76"},"agent_actions":{"view_html":"https://pith.science/pith/A7I7HD76JQAMWBIJX4MILLBYAG","download_json":"https://pith.science/pith/A7I7HD76JQAMWBIJX4MILLBYAG.json","view_paper":"https://pith.science/paper/A7I7HD76","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2404.03290&json=true","fetch_graph":"https://pith.science/api/pith-number/A7I7HD76JQAMWBIJX4MILLBYAG/graph.json","fetch_events":"https://pith.science/api/pith-number/A7I7HD76JQAMWBIJX4MILLBYAG/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/A7I7HD76JQAMWBIJX4MILLBYAG/action/timestamp_anchor","attest_storage":"https://pith.science/pith/A7I7HD76JQAMWBIJX4MILLBYAG/action/storage_attestation","attest_author":"https://pith.science/pith/A7I7HD76JQAMWBIJX4MILLBYAG/action/author_attestation","sign_citation":"https://pith.science/pith/A7I7HD76JQAMWBIJX4MILLBYAG/action/citation_signature","submit_replication":"https://pith.science/pith/A7I7HD76JQAMWBIJX4MILLBYAG/action/replication_record"}},"created_at":"2026-07-05T10:19:23.186471+00:00","updated_at":"2026-07-05T10:19:23.186471+00:00"}