{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:EXNW2WI4UTXEQAHSCMATF4JW3I","short_pith_number":"pith:EXNW2WI4","schema_version":"1.0","canonical_sha256":"25db6d591ca4ee4800f2130132f136da3f7cd679d759a69462133b3bacacd101","source":{"kind":"arxiv","id":"2206.14987","version":2},"attestation_state":"computed","paper":{"title":"Lookback for Learning to Branch","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["math.OC","stat.ML"],"primary_cat":"cs.LG","authors_text":"Andrea Lodi, Didier Chet\\'elat, Elias B. Khalil, Maxime Gasse, M. Pawan Kumar, Prateek Gupta, Yoshua Bengio","submitted_at":"2022-06-30T02:33:32Z","abstract_excerpt":"The expressive and computationally inexpensive bipartite Graph Neural Networks (GNN) have been shown to be an important component of deep learning based Mixed-Integer Linear Program (MILP) solvers. Recent works have demonstrated the effectiveness of such GNNs in replacing the branching (variable selection) heuristic in branch-and-bound (B&B) solvers. These GNNs are trained, offline and on a collection of MILPs, to imitate a very good but computationally expensive branching heuristic, strong branching. Given that B&B results in a tree of sub-MILPs, we ask (a) whether there are strong dependenci"},"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":"2206.14987","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2022-06-30T02:33:32Z","cross_cats_sorted":["math.OC","stat.ML"],"title_canon_sha256":"44139054a242b6102ab6015646c5b0f2c0df69c0184af135790a5bbe80c11284","abstract_canon_sha256":"a83086b3f192b5ebfbf513f4e76b9b9f9c2ece5e3b14c089a54ba595f8d3eb0c"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T05:29:02.535551Z","signature_b64":"0DB9j13ENiqZak7Fi6vbQBcYs5VLGiSXUBPOWlmgVbuE6n9dJhDiUDMymz5L4+tv1hlXeZDWhGkSHrNJdQD5DA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"25db6d591ca4ee4800f2130132f136da3f7cd679d759a69462133b3bacacd101","last_reissued_at":"2026-07-05T05:29:02.534959Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T05:29:02.534959Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Lookback for Learning to Branch","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["math.OC","stat.ML"],"primary_cat":"cs.LG","authors_text":"Andrea Lodi, Didier Chet\\'elat, Elias B. Khalil, Maxime Gasse, M. Pawan Kumar, Prateek Gupta, Yoshua Bengio","submitted_at":"2022-06-30T02:33:32Z","abstract_excerpt":"The expressive and computationally inexpensive bipartite Graph Neural Networks (GNN) have been shown to be an important component of deep learning based Mixed-Integer Linear Program (MILP) solvers. Recent works have demonstrated the effectiveness of such GNNs in replacing the branching (variable selection) heuristic in branch-and-bound (B&B) solvers. These GNNs are trained, offline and on a collection of MILPs, to imitate a very good but computationally expensive branching heuristic, strong branching. Given that B&B results in a tree of sub-MILPs, we ask (a) whether there are strong dependenci"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2206.14987","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/2206.14987/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":"2206.14987","created_at":"2026-07-05T05:29:02.535014+00:00"},{"alias_kind":"arxiv_version","alias_value":"2206.14987v2","created_at":"2026-07-05T05:29:02.535014+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2206.14987","created_at":"2026-07-05T05:29:02.535014+00:00"},{"alias_kind":"pith_short_12","alias_value":"EXNW2WI4UTXE","created_at":"2026-07-05T05:29:02.535014+00:00"},{"alias_kind":"pith_short_16","alias_value":"EXNW2WI4UTXEQAHS","created_at":"2026-07-05T05:29:02.535014+00:00"},{"alias_kind":"pith_short_8","alias_value":"EXNW2WI4","created_at":"2026-07-05T05:29:02.535014+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.21483","citing_title":"A note on the convergence guarantees of RLT-based algorithms for polynomial optimization","ref_index":83,"is_internal_anchor":false},{"citing_arxiv_id":"2604.22107","citing_title":"A Hybrid Reinforcement and Self-Supervised Learning Aided Benders Decomposition Algorithm","ref_index":9,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/EXNW2WI4UTXEQAHSCMATF4JW3I","json":"https://pith.science/pith/EXNW2WI4UTXEQAHSCMATF4JW3I.json","graph_json":"https://pith.science/api/pith-number/EXNW2WI4UTXEQAHSCMATF4JW3I/graph.json","events_json":"https://pith.science/api/pith-number/EXNW2WI4UTXEQAHSCMATF4JW3I/events.json","paper":"https://pith.science/paper/EXNW2WI4"},"agent_actions":{"view_html":"https://pith.science/pith/EXNW2WI4UTXEQAHSCMATF4JW3I","download_json":"https://pith.science/pith/EXNW2WI4UTXEQAHSCMATF4JW3I.json","view_paper":"https://pith.science/paper/EXNW2WI4","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2206.14987&json=true","fetch_graph":"https://pith.science/api/pith-number/EXNW2WI4UTXEQAHSCMATF4JW3I/graph.json","fetch_events":"https://pith.science/api/pith-number/EXNW2WI4UTXEQAHSCMATF4JW3I/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/EXNW2WI4UTXEQAHSCMATF4JW3I/action/timestamp_anchor","attest_storage":"https://pith.science/pith/EXNW2WI4UTXEQAHSCMATF4JW3I/action/storage_attestation","attest_author":"https://pith.science/pith/EXNW2WI4UTXEQAHSCMATF4JW3I/action/author_attestation","sign_citation":"https://pith.science/pith/EXNW2WI4UTXEQAHSCMATF4JW3I/action/citation_signature","submit_replication":"https://pith.science/pith/EXNW2WI4UTXEQAHSCMATF4JW3I/action/replication_record"}},"created_at":"2026-07-05T05:29:02.535014+00:00","updated_at":"2026-07-05T05:29:02.535014+00:00"}