{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:TVQM32XL4EUVIKUV4NVAT4NF26","short_pith_number":"pith:TVQM32XL","schema_version":"1.0","canonical_sha256":"9d60cdeaebe129542a95e36a09f1a5d7af324ef5fa544f8ead63de06b06dcf2a","source":{"kind":"arxiv","id":"2505.11342","version":1},"attestation_state":"computed","paper":{"title":"Sobolev Training of End-to-End Optimization Proxies","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["math.OC"],"primary_cat":"cs.LG","authors_text":"Andrew W. Rosemberg, Joaquim Dias Garcia, Pascal Van Hentenryck, Russell Bent","submitted_at":"2025-05-16T15:10:01Z","abstract_excerpt":"Optimization proxies - machine learning models trained to approximate the solution mapping of parametric optimization problems in a single forward pass - offer dramatic reductions in inference time compared to traditional iterative solvers. This work investigates the integration of solver sensitivities into such end to end proxies via a Sobolev training paradigm and does so in two distinct settings: (i) fully supervised proxies, where exact solver outputs and sensitivities are available, and (ii) self supervised proxies that rely only on the objective and constraint structure of the underlying"},"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":"2505.11342","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.LG","submitted_at":"2025-05-16T15:10:01Z","cross_cats_sorted":["math.OC"],"title_canon_sha256":"7eb4ba109e1bce3c13ccd44925b695b5696692ebb01495ef6dd15ddc8ca5774c","abstract_canon_sha256":"348764a765593189548d37f7bb90c61c481db4993b2c5a4ea11a786efd7df01a"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:04:14.677705Z","signature_b64":"AjbBhK68ScKSC9PqlDkzSEAYomO98KODjMhfx0FOfS89g4qfirCzIjv4hRpONC60pRfuIORx7zGAuEB6cA0cDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"9d60cdeaebe129542a95e36a09f1a5d7af324ef5fa544f8ead63de06b06dcf2a","last_reissued_at":"2026-07-05T11:04:14.677169Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:04:14.677169Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Sobolev Training of End-to-End Optimization Proxies","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["math.OC"],"primary_cat":"cs.LG","authors_text":"Andrew W. Rosemberg, Joaquim Dias Garcia, Pascal Van Hentenryck, Russell Bent","submitted_at":"2025-05-16T15:10:01Z","abstract_excerpt":"Optimization proxies - machine learning models trained to approximate the solution mapping of parametric optimization problems in a single forward pass - offer dramatic reductions in inference time compared to traditional iterative solvers. This work investigates the integration of solver sensitivities into such end to end proxies via a Sobolev training paradigm and does so in two distinct settings: (i) fully supervised proxies, where exact solver outputs and sensitivities are available, and (ii) self supervised proxies that rely only on the objective and constraint structure of the underlying"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.11342","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/2505.11342/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":"2505.11342","created_at":"2026-07-05T11:04:14.677246+00:00"},{"alias_kind":"arxiv_version","alias_value":"2505.11342v1","created_at":"2026-07-05T11:04:14.677246+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.11342","created_at":"2026-07-05T11:04:14.677246+00:00"},{"alias_kind":"pith_short_12","alias_value":"TVQM32XL4EUV","created_at":"2026-07-05T11:04:14.677246+00:00"},{"alias_kind":"pith_short_16","alias_value":"TVQM32XL4EUVIKUV","created_at":"2026-07-05T11:04:14.677246+00:00"},{"alias_kind":"pith_short_8","alias_value":"TVQM32XL","created_at":"2026-07-05T11:04:14.677246+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2604.22746","citing_title":"Relaxation-Informed Training of Neural Network Surrogate Models","ref_index":53,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/TVQM32XL4EUVIKUV4NVAT4NF26","json":"https://pith.science/pith/TVQM32XL4EUVIKUV4NVAT4NF26.json","graph_json":"https://pith.science/api/pith-number/TVQM32XL4EUVIKUV4NVAT4NF26/graph.json","events_json":"https://pith.science/api/pith-number/TVQM32XL4EUVIKUV4NVAT4NF26/events.json","paper":"https://pith.science/paper/TVQM32XL"},"agent_actions":{"view_html":"https://pith.science/pith/TVQM32XL4EUVIKUV4NVAT4NF26","download_json":"https://pith.science/pith/TVQM32XL4EUVIKUV4NVAT4NF26.json","view_paper":"https://pith.science/paper/TVQM32XL","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2505.11342&json=true","fetch_graph":"https://pith.science/api/pith-number/TVQM32XL4EUVIKUV4NVAT4NF26/graph.json","fetch_events":"https://pith.science/api/pith-number/TVQM32XL4EUVIKUV4NVAT4NF26/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/TVQM32XL4EUVIKUV4NVAT4NF26/action/timestamp_anchor","attest_storage":"https://pith.science/pith/TVQM32XL4EUVIKUV4NVAT4NF26/action/storage_attestation","attest_author":"https://pith.science/pith/TVQM32XL4EUVIKUV4NVAT4NF26/action/author_attestation","sign_citation":"https://pith.science/pith/TVQM32XL4EUVIKUV4NVAT4NF26/action/citation_signature","submit_replication":"https://pith.science/pith/TVQM32XL4EUVIKUV4NVAT4NF26/action/replication_record"}},"created_at":"2026-07-05T11:04:14.677246+00:00","updated_at":"2026-07-05T11:04:14.677246+00:00"}