{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2017:2HTWYUHARE3PZU2QYOA2G3KOQI","short_pith_number":"pith:2HTWYUHA","schema_version":"1.0","canonical_sha256":"d1e76c50e08936fcd350c381a36d4e8217261ab26d55d54df66cef218b841804","source":{"kind":"arxiv","id":"1710.08005","version":5},"attestation_state":"computed","paper":{"title":"Smart \"Predict, then Optimize\"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG","stat.ML"],"primary_cat":"math.OC","authors_text":"Adam N. Elmachtoub, Paul Grigas","submitted_at":"2017-10-22T19:44:46Z","abstract_excerpt":"Many real-world analytics problems involve two significant challenges: prediction and optimization. Due to the typically complex nature of each challenge, the standard paradigm is predict-then-optimize. By and large, machine learning tools are intended to minimize prediction error and do not account for how the predictions will be used in the downstream optimization problem. In contrast, we propose a new and very general framework, called Smart \"Predict, then Optimize\" (SPO), which directly leverages the optimization problem structure, i.e., its objective and constraints, for designing better "},"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":"1710.08005","kind":"arxiv","version":5},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"math.OC","submitted_at":"2017-10-22T19:44:46Z","cross_cats_sorted":["cs.LG","stat.ML"],"title_canon_sha256":"e0e6d413f474d36b464105dffd33872e86e86160d7a0fc918f2ae30b126c8f41","abstract_canon_sha256":"c9651bee8929412e86372d316354724b052ff42ebc7fa83e2fe86b8e239bac2e"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T01:53:05.472180Z","signature_b64":"X4V5ju3udaBwterc6666SKRQluHUMK6osBWWE2eomeq6shjFG4/Q4v+on1AUyTDQyG3K9un4ed2evX1VUSjXCg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"d1e76c50e08936fcd350c381a36d4e8217261ab26d55d54df66cef218b841804","last_reissued_at":"2026-07-05T01:53:05.471702Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T01:53:05.471702Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Smart \"Predict, then Optimize\"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG","stat.ML"],"primary_cat":"math.OC","authors_text":"Adam N. Elmachtoub, Paul Grigas","submitted_at":"2017-10-22T19:44:46Z","abstract_excerpt":"Many real-world analytics problems involve two significant challenges: prediction and optimization. Due to the typically complex nature of each challenge, the standard paradigm is predict-then-optimize. By and large, machine learning tools are intended to minimize prediction error and do not account for how the predictions will be used in the downstream optimization problem. In contrast, we propose a new and very general framework, called Smart \"Predict, then Optimize\" (SPO), which directly leverages the optimization problem structure, i.e., its objective and constraints, for designing better "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1710.08005","kind":"arxiv","version":5},"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/1710.08005/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":"1710.08005","created_at":"2026-07-05T01:53:05.471771+00:00"},{"alias_kind":"arxiv_version","alias_value":"1710.08005v5","created_at":"2026-07-05T01:53:05.471771+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1710.08005","created_at":"2026-07-05T01:53:05.471771+00:00"},{"alias_kind":"pith_short_12","alias_value":"2HTWYUHARE3P","created_at":"2026-07-05T01:53:05.471771+00:00"},{"alias_kind":"pith_short_16","alias_value":"2HTWYUHARE3PZU2Q","created_at":"2026-07-05T01:53:05.471771+00:00"},{"alias_kind":"pith_short_8","alias_value":"2HTWYUHA","created_at":"2026-07-05T01:53:05.471771+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":4,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.25274","citing_title":"UC-Search: Risk-Aware Test-Time Search for Delayed Constrained Time-Series Control","ref_index":61,"is_internal_anchor":false},{"citing_arxiv_id":"2606.24996","citing_title":"From Forecasting Leaderboards to Deployment Decisions: A Fail-Closed Certification Protocol","ref_index":16,"is_internal_anchor":false},{"citing_arxiv_id":"1907.05912","citing_title":"MIPaaL: Mixed Integer Program as a Layer","ref_index":16,"is_internal_anchor":false},{"citing_arxiv_id":"2604.11507","citing_title":"Deep Learning for Sequential Decision Making under Uncertainty: Foundations, Frameworks, and Frontiers","ref_index":42,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/2HTWYUHARE3PZU2QYOA2G3KOQI","json":"https://pith.science/pith/2HTWYUHARE3PZU2QYOA2G3KOQI.json","graph_json":"https://pith.science/api/pith-number/2HTWYUHARE3PZU2QYOA2G3KOQI/graph.json","events_json":"https://pith.science/api/pith-number/2HTWYUHARE3PZU2QYOA2G3KOQI/events.json","paper":"https://pith.science/paper/2HTWYUHA"},"agent_actions":{"view_html":"https://pith.science/pith/2HTWYUHARE3PZU2QYOA2G3KOQI","download_json":"https://pith.science/pith/2HTWYUHARE3PZU2QYOA2G3KOQI.json","view_paper":"https://pith.science/paper/2HTWYUHA","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1710.08005&json=true","fetch_graph":"https://pith.science/api/pith-number/2HTWYUHARE3PZU2QYOA2G3KOQI/graph.json","fetch_events":"https://pith.science/api/pith-number/2HTWYUHARE3PZU2QYOA2G3KOQI/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/2HTWYUHARE3PZU2QYOA2G3KOQI/action/timestamp_anchor","attest_storage":"https://pith.science/pith/2HTWYUHARE3PZU2QYOA2G3KOQI/action/storage_attestation","attest_author":"https://pith.science/pith/2HTWYUHARE3PZU2QYOA2G3KOQI/action/author_attestation","sign_citation":"https://pith.science/pith/2HTWYUHARE3PZU2QYOA2G3KOQI/action/citation_signature","submit_replication":"https://pith.science/pith/2HTWYUHARE3PZU2QYOA2G3KOQI/action/replication_record"}},"created_at":"2026-07-05T01:53:05.471771+00:00","updated_at":"2026-07-05T01:53:05.471771+00:00"}