{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:TALM4XO5Y6C6SQGCXNCJCOQCWT","short_pith_number":"pith:TALM4XO5","schema_version":"1.0","canonical_sha256":"9816ce5dddc785e940c2bb44913a02b4d53bbc9a9b824752e93bf41954297b29","source":{"kind":"arxiv","id":"2107.12855","version":1},"attestation_state":"computed","paper":{"title":"Neural Network Branch-and-Bound for Neural Network Verification","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Florian Jaeckle, Jingyue Lu, M. Pawan Kumar","submitted_at":"2021-07-27T14:42:57Z","abstract_excerpt":"Many available formal verification methods have been shown to be instances of a unified Branch-and-Bound (BaB) formulation. We propose a novel machine learning framework that can be used for designing an effective branching strategy as well as for computing better lower bounds. Specifically, we learn two graph neural networks (GNN) that both directly treat the network we want to verify as a graph input and perform forward-backward passes through the GNN layers. We use one GNN to simulate the strong branching heuristic behaviour and another to compute a feasible dual solution of the convex rela"},"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":"2107.12855","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.LG","submitted_at":"2021-07-27T14:42:57Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"03f6a005fcaec4bcc71808f892c8f94197ffe95f05587e564ae0c5cf574a78d5","abstract_canon_sha256":"64541231f424b0bc59a16e624fe1395f98cb6741551bb1cde65f2bf2a0605f33"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T03:01:01.420008Z","signature_b64":"LZJ3VkQrOLMTBCqrG5EbLHvN992M5Zt3m6fQYJpPBNeC8o6VHFyhjr94iaH+kf9IRGrdpFluLfzjpbGu/3b4Ag==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"9816ce5dddc785e940c2bb44913a02b4d53bbc9a9b824752e93bf41954297b29","last_reissued_at":"2026-07-05T03:01:01.419496Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T03:01:01.419496Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Neural Network Branch-and-Bound for Neural Network Verification","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Florian Jaeckle, Jingyue Lu, M. Pawan Kumar","submitted_at":"2021-07-27T14:42:57Z","abstract_excerpt":"Many available formal verification methods have been shown to be instances of a unified Branch-and-Bound (BaB) formulation. We propose a novel machine learning framework that can be used for designing an effective branching strategy as well as for computing better lower bounds. Specifically, we learn two graph neural networks (GNN) that both directly treat the network we want to verify as a graph input and perform forward-backward passes through the GNN layers. We use one GNN to simulate the strong branching heuristic behaviour and another to compute a feasible dual solution of the convex rela"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2107.12855","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/2107.12855/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":"2107.12855","created_at":"2026-07-05T03:01:01.419558+00:00"},{"alias_kind":"arxiv_version","alias_value":"2107.12855v1","created_at":"2026-07-05T03:01:01.419558+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2107.12855","created_at":"2026-07-05T03:01:01.419558+00:00"},{"alias_kind":"pith_short_12","alias_value":"TALM4XO5Y6C6","created_at":"2026-07-05T03:01:01.419558+00:00"},{"alias_kind":"pith_short_16","alias_value":"TALM4XO5Y6C6SQGC","created_at":"2026-07-05T03:01:01.419558+00:00"},{"alias_kind":"pith_short_8","alias_value":"TALM4XO5","created_at":"2026-07-05T03:01:01.419558+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2412.03028","citing_title":"Specification Generation for Neural Networks in Systems","ref_index":33,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/TALM4XO5Y6C6SQGCXNCJCOQCWT","json":"https://pith.science/pith/TALM4XO5Y6C6SQGCXNCJCOQCWT.json","graph_json":"https://pith.science/api/pith-number/TALM4XO5Y6C6SQGCXNCJCOQCWT/graph.json","events_json":"https://pith.science/api/pith-number/TALM4XO5Y6C6SQGCXNCJCOQCWT/events.json","paper":"https://pith.science/paper/TALM4XO5"},"agent_actions":{"view_html":"https://pith.science/pith/TALM4XO5Y6C6SQGCXNCJCOQCWT","download_json":"https://pith.science/pith/TALM4XO5Y6C6SQGCXNCJCOQCWT.json","view_paper":"https://pith.science/paper/TALM4XO5","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2107.12855&json=true","fetch_graph":"https://pith.science/api/pith-number/TALM4XO5Y6C6SQGCXNCJCOQCWT/graph.json","fetch_events":"https://pith.science/api/pith-number/TALM4XO5Y6C6SQGCXNCJCOQCWT/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/TALM4XO5Y6C6SQGCXNCJCOQCWT/action/timestamp_anchor","attest_storage":"https://pith.science/pith/TALM4XO5Y6C6SQGCXNCJCOQCWT/action/storage_attestation","attest_author":"https://pith.science/pith/TALM4XO5Y6C6SQGCXNCJCOQCWT/action/author_attestation","sign_citation":"https://pith.science/pith/TALM4XO5Y6C6SQGCXNCJCOQCWT/action/citation_signature","submit_replication":"https://pith.science/pith/TALM4XO5Y6C6SQGCXNCJCOQCWT/action/replication_record"}},"created_at":"2026-07-05T03:01:01.419558+00:00","updated_at":"2026-07-05T03:01:01.419558+00:00"}