{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:IYQBYXI2G5NHUHJISUDEE3FL2M","short_pith_number":"pith:IYQBYXI2","schema_version":"1.0","canonical_sha256":"46201c5d1a375a7a1d289506426cabd333814ba3928c2345cb9902a3e73653a7","source":{"kind":"arxiv","id":"2104.12225","version":1},"attestation_state":"computed","paper":{"title":"DC3: A learning method for optimization with hard constraints","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["math.OC","stat.ML"],"primary_cat":"cs.LG","authors_text":"David Rolnick, J. Zico Kolter, Priya L. Donti","submitted_at":"2021-04-25T18:21:59Z","abstract_excerpt":"Large optimization problems with hard constraints arise in many settings, yet classical solvers are often prohibitively slow, motivating the use of deep networks as cheap \"approximate solvers.\" Unfortunately, naive deep learning approaches typically cannot enforce the hard constraints of such problems, leading to infeasible solutions. In this work, we present Deep Constraint Completion and Correction (DC3), an algorithm to address this challenge. Specifically, this method enforces feasibility via a differentiable procedure, which implicitly completes partial solutions to satisfy equality const"},"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":"2104.12225","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2021-04-25T18:21:59Z","cross_cats_sorted":["math.OC","stat.ML"],"title_canon_sha256":"0c15ce3491daffbff19d770e901ac7173a613ac14d4a7ccf36fd068bdf022e8c","abstract_canon_sha256":"02c7760975b11fe8b4a2927fda4514b16696c104f9e398a03d082c5392ab7234"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T02:34:54.619248Z","signature_b64":"SgvLrzuwJx2RLSEl3TQatSfGduO+8labHz3JfboPQUvHbkCBPQcY0UpoVuh34Iv9BV1UsSTA4D5I9TH6tNNKDg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"46201c5d1a375a7a1d289506426cabd333814ba3928c2345cb9902a3e73653a7","last_reissued_at":"2026-07-05T02:34:54.618802Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T02:34:54.618802Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"DC3: A learning method for optimization with hard constraints","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["math.OC","stat.ML"],"primary_cat":"cs.LG","authors_text":"David Rolnick, J. Zico Kolter, Priya L. Donti","submitted_at":"2021-04-25T18:21:59Z","abstract_excerpt":"Large optimization problems with hard constraints arise in many settings, yet classical solvers are often prohibitively slow, motivating the use of deep networks as cheap \"approximate solvers.\" Unfortunately, naive deep learning approaches typically cannot enforce the hard constraints of such problems, leading to infeasible solutions. In this work, we present Deep Constraint Completion and Correction (DC3), an algorithm to address this challenge. Specifically, this method enforces feasibility via a differentiable procedure, which implicitly completes partial solutions to satisfy equality const"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2104.12225","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/2104.12225/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":"2104.12225","created_at":"2026-07-05T02:34:54.618856+00:00"},{"alias_kind":"arxiv_version","alias_value":"2104.12225v1","created_at":"2026-07-05T02:34:54.618856+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2104.12225","created_at":"2026-07-05T02:34:54.618856+00:00"},{"alias_kind":"pith_short_12","alias_value":"IYQBYXI2G5NH","created_at":"2026-07-05T02:34:54.618856+00:00"},{"alias_kind":"pith_short_16","alias_value":"IYQBYXI2G5NHUHJI","created_at":"2026-07-05T02:34:54.618856+00:00"},{"alias_kind":"pith_short_8","alias_value":"IYQBYXI2","created_at":"2026-07-05T02:34:54.618856+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":8,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.21728","citing_title":"Embedding Linear Equality Constraints in Probabilistic Neural Networks for Dynamic Modelling","ref_index":8,"is_internal_anchor":false},{"citing_arxiv_id":"2606.21440","citing_title":"Fusing Backdoors, Machine Learning, and Optimization for Large-Scale Parametric Mixed-Integer Programs","ref_index":14,"is_internal_anchor":false},{"citing_arxiv_id":"2606.05772","citing_title":"Physics-Informed Graph Learning Acceleration for Large-Scale AC-OPF with Topology Changes","ref_index":46,"is_internal_anchor":false},{"citing_arxiv_id":"2606.28817","citing_title":"Trust-Calibrated Certified Repair for Physics-Constrained Decisions under Localized Model Misspecification","ref_index":5,"is_internal_anchor":false},{"citing_arxiv_id":"2307.08336","citing_title":"RAYEN: Imposition of Hard Convex Constraints on Neural Networks","ref_index":24,"is_internal_anchor":false},{"citing_arxiv_id":"2602.09317","citing_title":"SnareNet: Flexible Repair Layers for Neural Networks with Hard Constraints","ref_index":10,"is_internal_anchor":false},{"citing_arxiv_id":"2605.14019","citing_title":"Regret Equals Covariance: A Closed-Form Characterization for Stochastic Optimization","ref_index":8,"is_internal_anchor":false},{"citing_arxiv_id":"2604.24911","citing_title":"Learning with Embedded Linear Equality Constraints via Variational Bayesian Inference","ref_index":2,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/IYQBYXI2G5NHUHJISUDEE3FL2M","json":"https://pith.science/pith/IYQBYXI2G5NHUHJISUDEE3FL2M.json","graph_json":"https://pith.science/api/pith-number/IYQBYXI2G5NHUHJISUDEE3FL2M/graph.json","events_json":"https://pith.science/api/pith-number/IYQBYXI2G5NHUHJISUDEE3FL2M/events.json","paper":"https://pith.science/paper/IYQBYXI2"},"agent_actions":{"view_html":"https://pith.science/pith/IYQBYXI2G5NHUHJISUDEE3FL2M","download_json":"https://pith.science/pith/IYQBYXI2G5NHUHJISUDEE3FL2M.json","view_paper":"https://pith.science/paper/IYQBYXI2","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2104.12225&json=true","fetch_graph":"https://pith.science/api/pith-number/IYQBYXI2G5NHUHJISUDEE3FL2M/graph.json","fetch_events":"https://pith.science/api/pith-number/IYQBYXI2G5NHUHJISUDEE3FL2M/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/IYQBYXI2G5NHUHJISUDEE3FL2M/action/timestamp_anchor","attest_storage":"https://pith.science/pith/IYQBYXI2G5NHUHJISUDEE3FL2M/action/storage_attestation","attest_author":"https://pith.science/pith/IYQBYXI2G5NHUHJISUDEE3FL2M/action/author_attestation","sign_citation":"https://pith.science/pith/IYQBYXI2G5NHUHJISUDEE3FL2M/action/citation_signature","submit_replication":"https://pith.science/pith/IYQBYXI2G5NHUHJISUDEE3FL2M/action/replication_record"}},"created_at":"2026-07-05T02:34:54.618856+00:00","updated_at":"2026-07-05T02:34:54.618856+00:00"}