{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:W6HBGO5YIJASZAZRUNNLBW5JMD","short_pith_number":"pith:W6HBGO5Y","schema_version":"1.0","canonical_sha256":"b78e133bb842412c8331a35ab0dba960d0278a1d15e8f28e2a8f92d8ae2d9ed6","source":{"kind":"arxiv","id":"2411.09117","version":1},"attestation_state":"computed","paper":{"title":"Efficiently learning and sampling multimodal distributions with data-based initialization","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.DS","math.PR","stat.ML"],"primary_cat":"cs.LG","authors_text":"Frederic Koehler, Holden Lee, Thuy-Duong Vuong","submitted_at":"2024-11-14T01:37:02Z","abstract_excerpt":"We consider the problem of sampling a multimodal distribution with a Markov chain given a small number of samples from the stationary measure. Although mixing can be arbitrarily slow, we show that if the Markov chain has a $k$th order spectral gap, initialization from a set of $\\tilde O(k/\\varepsilon^2)$ samples from the stationary distribution will, with high probability over the samples, efficiently generate a sample whose conditional law is $\\varepsilon$-close in TV distance to the stationary measure. In particular, this applies to mixtures of $k$ distributions satisfying a Poincar\\'e inequ"},"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":"2411.09117","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2024-11-14T01:37:02Z","cross_cats_sorted":["cs.DS","math.PR","stat.ML"],"title_canon_sha256":"9fe8643d7a97a35d6b701815ed07a074b4410e20974d2a898879e638a75f9cda","abstract_canon_sha256":"23faf997c4f6974a233f37698adcff3b34a009421191a7e2b7cd0b3f389e6ef5"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:35:20.402747Z","signature_b64":"RGvbE/4pZSRsWmJI7y7m4nq+3XTJiglOZZWDoQc4kakQ32NHh2qnP+9vyaIpFGUlHLfkJ33GrWOB0AkBLvb+Ag==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"b78e133bb842412c8331a35ab0dba960d0278a1d15e8f28e2a8f92d8ae2d9ed6","last_reissued_at":"2026-07-05T09:35:20.402311Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:35:20.402311Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Efficiently learning and sampling multimodal distributions with data-based initialization","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.DS","math.PR","stat.ML"],"primary_cat":"cs.LG","authors_text":"Frederic Koehler, Holden Lee, Thuy-Duong Vuong","submitted_at":"2024-11-14T01:37:02Z","abstract_excerpt":"We consider the problem of sampling a multimodal distribution with a Markov chain given a small number of samples from the stationary measure. Although mixing can be arbitrarily slow, we show that if the Markov chain has a $k$th order spectral gap, initialization from a set of $\\tilde O(k/\\varepsilon^2)$ samples from the stationary distribution will, with high probability over the samples, efficiently generate a sample whose conditional law is $\\varepsilon$-close in TV distance to the stationary measure. In particular, this applies to mixtures of $k$ distributions satisfying a Poincar\\'e inequ"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2411.09117","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/2411.09117/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":"2411.09117","created_at":"2026-07-05T09:35:20.402387+00:00"},{"alias_kind":"arxiv_version","alias_value":"2411.09117v1","created_at":"2026-07-05T09:35:20.402387+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2411.09117","created_at":"2026-07-05T09:35:20.402387+00:00"},{"alias_kind":"pith_short_12","alias_value":"W6HBGO5YIJAS","created_at":"2026-07-05T09:35:20.402387+00:00"},{"alias_kind":"pith_short_16","alias_value":"W6HBGO5YIJASZAZR","created_at":"2026-07-05T09:35:20.402387+00:00"},{"alias_kind":"pith_short_8","alias_value":"W6HBGO5Y","created_at":"2026-07-05T09:35:20.402387+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.24752","citing_title":"A computational phase transition for learning-to-sample from Ising models","ref_index":3,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/W6HBGO5YIJASZAZRUNNLBW5JMD","json":"https://pith.science/pith/W6HBGO5YIJASZAZRUNNLBW5JMD.json","graph_json":"https://pith.science/api/pith-number/W6HBGO5YIJASZAZRUNNLBW5JMD/graph.json","events_json":"https://pith.science/api/pith-number/W6HBGO5YIJASZAZRUNNLBW5JMD/events.json","paper":"https://pith.science/paper/W6HBGO5Y"},"agent_actions":{"view_html":"https://pith.science/pith/W6HBGO5YIJASZAZRUNNLBW5JMD","download_json":"https://pith.science/pith/W6HBGO5YIJASZAZRUNNLBW5JMD.json","view_paper":"https://pith.science/paper/W6HBGO5Y","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2411.09117&json=true","fetch_graph":"https://pith.science/api/pith-number/W6HBGO5YIJASZAZRUNNLBW5JMD/graph.json","fetch_events":"https://pith.science/api/pith-number/W6HBGO5YIJASZAZRUNNLBW5JMD/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/W6HBGO5YIJASZAZRUNNLBW5JMD/action/timestamp_anchor","attest_storage":"https://pith.science/pith/W6HBGO5YIJASZAZRUNNLBW5JMD/action/storage_attestation","attest_author":"https://pith.science/pith/W6HBGO5YIJASZAZRUNNLBW5JMD/action/author_attestation","sign_citation":"https://pith.science/pith/W6HBGO5YIJASZAZRUNNLBW5JMD/action/citation_signature","submit_replication":"https://pith.science/pith/W6HBGO5YIJASZAZRUNNLBW5JMD/action/replication_record"}},"created_at":"2026-07-05T09:35:20.402387+00:00","updated_at":"2026-07-05T09:35:20.402387+00:00"}