{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2015:YQK3T5PBX3S7OUWWKIRZQSXHYG","short_pith_number":"pith:YQK3T5PB","schema_version":"1.0","canonical_sha256":"c415b9f5e1bee5f752d65223984ae7c18679ff910f6a73bf4d1b5ecdddb0b05d","source":{"kind":"arxiv","id":"1511.02124","version":2},"attestation_state":"computed","paper":{"title":"Barrier Frank-Wolfe for Marginal Inference","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG","math.OC"],"primary_cat":"stat.ML","authors_text":"David Sontag, Rahul G. Krishnan, Simon Lacoste-Julien","submitted_at":"2015-11-06T15:48:53Z","abstract_excerpt":"We introduce a globally-convergent algorithm for optimizing the tree-reweighted (TRW) variational objective over the marginal polytope. The algorithm is based on the conditional gradient method (Frank-Wolfe) and moves pseudomarginals within the marginal polytope through repeated maximum a posteriori (MAP) calls. This modular structure enables us to leverage black-box MAP solvers (both exact and approximate) for variational inference, and obtains more accurate results than tree-reweighted algorithms that optimize over the local consistency relaxation. Theoretically, we bound the sub-optimality "},"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":"1511.02124","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"stat.ML","submitted_at":"2015-11-06T15:48:53Z","cross_cats_sorted":["cs.LG","math.OC"],"title_canon_sha256":"4ffa290c25fd2adba613563d102c7759f711ad33616dd3ab05c4facb87feb386","abstract_canon_sha256":"db71f8b716f7e04b776fdd344da6c1bbb4cd10746cf58c1b8ef0934c07c75582"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-18T01:26:02.432390Z","signature_b64":"ihfNzFGf+NJ49JbzbJ0HnE1vYZ8AKeZDysjEAk/fn06Hxvhpo6PO//Bmn8tmQPQgGfl2HLVMvy+bia9+yOBcDw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"c415b9f5e1bee5f752d65223984ae7c18679ff910f6a73bf4d1b5ecdddb0b05d","last_reissued_at":"2026-05-18T01:26:02.431910Z","signature_status":"signed_v1","first_computed_at":"2026-05-18T01:26:02.431910Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Barrier Frank-Wolfe for Marginal Inference","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG","math.OC"],"primary_cat":"stat.ML","authors_text":"David Sontag, Rahul G. Krishnan, Simon Lacoste-Julien","submitted_at":"2015-11-06T15:48:53Z","abstract_excerpt":"We introduce a globally-convergent algorithm for optimizing the tree-reweighted (TRW) variational objective over the marginal polytope. The algorithm is based on the conditional gradient method (Frank-Wolfe) and moves pseudomarginals within the marginal polytope through repeated maximum a posteriori (MAP) calls. This modular structure enables us to leverage black-box MAP solvers (both exact and approximate) for variational inference, and obtains more accurate results than tree-reweighted algorithms that optimize over the local consistency relaxation. Theoretically, we bound the sub-optimality "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1511.02124","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":""},"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":"1511.02124","created_at":"2026-05-18T01:26:02.431983+00:00"},{"alias_kind":"arxiv_version","alias_value":"1511.02124v2","created_at":"2026-05-18T01:26:02.431983+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1511.02124","created_at":"2026-05-18T01:26:02.431983+00:00"},{"alias_kind":"pith_short_12","alias_value":"YQK3T5PBX3S7","created_at":"2026-05-18T12:29:50.041715+00:00"},{"alias_kind":"pith_short_16","alias_value":"YQK3T5PBX3S7OUWW","created_at":"2026-05-18T12:29:50.041715+00:00"},{"alias_kind":"pith_short_8","alias_value":"YQK3T5PB","created_at":"2026-05-18T12:29:50.041715+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/YQK3T5PBX3S7OUWWKIRZQSXHYG","json":"https://pith.science/pith/YQK3T5PBX3S7OUWWKIRZQSXHYG.json","graph_json":"https://pith.science/api/pith-number/YQK3T5PBX3S7OUWWKIRZQSXHYG/graph.json","events_json":"https://pith.science/api/pith-number/YQK3T5PBX3S7OUWWKIRZQSXHYG/events.json","paper":"https://pith.science/paper/YQK3T5PB"},"agent_actions":{"view_html":"https://pith.science/pith/YQK3T5PBX3S7OUWWKIRZQSXHYG","download_json":"https://pith.science/pith/YQK3T5PBX3S7OUWWKIRZQSXHYG.json","view_paper":"https://pith.science/paper/YQK3T5PB","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1511.02124&json=true","fetch_graph":"https://pith.science/api/pith-number/YQK3T5PBX3S7OUWWKIRZQSXHYG/graph.json","fetch_events":"https://pith.science/api/pith-number/YQK3T5PBX3S7OUWWKIRZQSXHYG/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/YQK3T5PBX3S7OUWWKIRZQSXHYG/action/timestamp_anchor","attest_storage":"https://pith.science/pith/YQK3T5PBX3S7OUWWKIRZQSXHYG/action/storage_attestation","attest_author":"https://pith.science/pith/YQK3T5PBX3S7OUWWKIRZQSXHYG/action/author_attestation","sign_citation":"https://pith.science/pith/YQK3T5PBX3S7OUWWKIRZQSXHYG/action/citation_signature","submit_replication":"https://pith.science/pith/YQK3T5PBX3S7OUWWKIRZQSXHYG/action/replication_record"}},"created_at":"2026-05-18T01:26:02.431983+00:00","updated_at":"2026-05-18T01:26:02.431983+00:00"}