{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:2UZIY3H2GTTVFBSM7WWNNCS7AV","short_pith_number":"pith:2UZIY3H2","schema_version":"1.0","canonical_sha256":"d5328c6cfa34e752864cfdacd68a5f056502018c12f3a8845d2fea82a7d62953","source":{"kind":"arxiv","id":"2201.13117","version":3},"attestation_state":"computed","paper":{"title":"Continual Repeated Annealed Flow Transport Monte Carlo","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cond-mat.stat-mech","cs.LG","hep-lat"],"primary_cat":"stat.ML","authors_text":"Alexander G. D. G. Matthews, Arnaud Doucet, Danilo J. Rezende, Michael Arbel","submitted_at":"2022-01-31T10:58:31Z","abstract_excerpt":"We propose Continual Repeated Annealed Flow Transport Monte Carlo (CRAFT), a method that combines a sequential Monte Carlo (SMC) sampler (itself a generalization of Annealed Importance Sampling) with variational inference using normalizing flows. The normalizing flows are directly trained to transport between annealing temperatures using a KL divergence for each transition. This optimization objective is itself estimated using the normalizing flow/SMC approximation. We show conceptually and using multiple empirical examples that CRAFT improves on Annealed Flow Transport Monte Carlo (Arbel et a"},"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":"2201.13117","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"stat.ML","submitted_at":"2022-01-31T10:58:31Z","cross_cats_sorted":["cond-mat.stat-mech","cs.LG","hep-lat"],"title_canon_sha256":"87254c8ef02266c5e88541fed3222ac1830321750e8bcca01766001198f48883","abstract_canon_sha256":"a20dab541b5d5c42ba6e69c38a74699531423bc63e2c5a8b85dd1b007e64befe"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T05:58:37.804247Z","signature_b64":"t+DDIpAMmIj/bSQHp09RjIlgjFb0v6XHs5Tfct60WVVjKSUEkpWKAko9vaCiT7o7KxCiQBALy12Kcn7Fu8vHBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"d5328c6cfa34e752864cfdacd68a5f056502018c12f3a8845d2fea82a7d62953","last_reissued_at":"2026-07-05T05:58:37.803749Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T05:58:37.803749Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Continual Repeated Annealed Flow Transport Monte Carlo","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cond-mat.stat-mech","cs.LG","hep-lat"],"primary_cat":"stat.ML","authors_text":"Alexander G. D. G. Matthews, Arnaud Doucet, Danilo J. Rezende, Michael Arbel","submitted_at":"2022-01-31T10:58:31Z","abstract_excerpt":"We propose Continual Repeated Annealed Flow Transport Monte Carlo (CRAFT), a method that combines a sequential Monte Carlo (SMC) sampler (itself a generalization of Annealed Importance Sampling) with variational inference using normalizing flows. The normalizing flows are directly trained to transport between annealing temperatures using a KL divergence for each transition. This optimization objective is itself estimated using the normalizing flow/SMC approximation. We show conceptually and using multiple empirical examples that CRAFT improves on Annealed Flow Transport Monte Carlo (Arbel et a"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2201.13117","kind":"arxiv","version":3},"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/2201.13117/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":"2201.13117","created_at":"2026-07-05T05:58:37.803808+00:00"},{"alias_kind":"arxiv_version","alias_value":"2201.13117v3","created_at":"2026-07-05T05:58:37.803808+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2201.13117","created_at":"2026-07-05T05:58:37.803808+00:00"},{"alias_kind":"pith_short_12","alias_value":"2UZIY3H2GTTV","created_at":"2026-07-05T05:58:37.803808+00:00"},{"alias_kind":"pith_short_16","alias_value":"2UZIY3H2GTTVFBSM","created_at":"2026-07-05T05:58:37.803808+00:00"},{"alias_kind":"pith_short_8","alias_value":"2UZIY3H2","created_at":"2026-07-05T05:58:37.803808+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2510.25704","citing_title":"Scaling flow-based approaches for topology sampling in $\\mathrm{SU}(3)$ gauge theory","ref_index":102,"is_internal_anchor":false},{"citing_arxiv_id":"2604.05303","citing_title":"Jeffreys Flow: Robust Boltzmann Generators for Rare Event Sampling via Parallel Tempering Distillation","ref_index":52,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/2UZIY3H2GTTVFBSM7WWNNCS7AV","json":"https://pith.science/pith/2UZIY3H2GTTVFBSM7WWNNCS7AV.json","graph_json":"https://pith.science/api/pith-number/2UZIY3H2GTTVFBSM7WWNNCS7AV/graph.json","events_json":"https://pith.science/api/pith-number/2UZIY3H2GTTVFBSM7WWNNCS7AV/events.json","paper":"https://pith.science/paper/2UZIY3H2"},"agent_actions":{"view_html":"https://pith.science/pith/2UZIY3H2GTTVFBSM7WWNNCS7AV","download_json":"https://pith.science/pith/2UZIY3H2GTTVFBSM7WWNNCS7AV.json","view_paper":"https://pith.science/paper/2UZIY3H2","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2201.13117&json=true","fetch_graph":"https://pith.science/api/pith-number/2UZIY3H2GTTVFBSM7WWNNCS7AV/graph.json","fetch_events":"https://pith.science/api/pith-number/2UZIY3H2GTTVFBSM7WWNNCS7AV/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/2UZIY3H2GTTVFBSM7WWNNCS7AV/action/timestamp_anchor","attest_storage":"https://pith.science/pith/2UZIY3H2GTTVFBSM7WWNNCS7AV/action/storage_attestation","attest_author":"https://pith.science/pith/2UZIY3H2GTTVFBSM7WWNNCS7AV/action/author_attestation","sign_citation":"https://pith.science/pith/2UZIY3H2GTTVFBSM7WWNNCS7AV/action/citation_signature","submit_replication":"https://pith.science/pith/2UZIY3H2GTTVFBSM7WWNNCS7AV/action/replication_record"}},"created_at":"2026-07-05T05:58:37.803808+00:00","updated_at":"2026-07-05T05:58:37.803808+00:00"}