{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2020:4HBCCS2OAGID2O2MYXTXRLRE7P","short_pith_number":"pith:4HBCCS2O","schema_version":"1.0","canonical_sha256":"e1c2214b4e01903d3b4cc5e778ae24fbe70b30807e9621a764d847dc49b1af50","source":{"kind":"arxiv","id":"2008.10087","version":3},"attestation_state":"computed","paper":{"title":"Blindness of score-based methods to isolated components and mixing proportions","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"stat.ML","authors_text":"Heishiro Kanagawa, Li K. Wenliang","submitted_at":"2020-08-23T18:24:42Z","abstract_excerpt":"Statistical tasks such as density estimation and approximate Bayesian inference often involve densities with unknown normalising constants. Score-based methods, including score matching, are popular techniques as they are free of normalising constants. Although these methods enjoy theoretical guarantees, a little-known fact is that they exhibit practical failure modes when the unnormalised distribution of interest has isolated components -- they cannot discover isolated components or identify the correct mixing proportions between components. We demonstrate these findings using simple distribu"},"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":"2008.10087","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"stat.ML","submitted_at":"2020-08-23T18:24:42Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"74f6b65a814576014fbbacd3e0ce60aa7d751ed4936eed4cedbb89b1dd42fa68","abstract_canon_sha256":"4d981c79cf37c0de6eb5a60192019a6b6a8e9b0bec725d00614367347e31643b"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T03:42:19.676257Z","signature_b64":"2/MsYk4IUo7O3lsai4ZfH+7JWo2t35K8YtyxnEWq8Ojboa0j37Ci9lVeJUTXs9Hzynv2+RRx7wvBngCoosi1Aw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"e1c2214b4e01903d3b4cc5e778ae24fbe70b30807e9621a764d847dc49b1af50","last_reissued_at":"2026-07-05T03:42:19.675767Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T03:42:19.675767Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Blindness of score-based methods to isolated components and mixing proportions","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"stat.ML","authors_text":"Heishiro Kanagawa, Li K. Wenliang","submitted_at":"2020-08-23T18:24:42Z","abstract_excerpt":"Statistical tasks such as density estimation and approximate Bayesian inference often involve densities with unknown normalising constants. Score-based methods, including score matching, are popular techniques as they are free of normalising constants. Although these methods enjoy theoretical guarantees, a little-known fact is that they exhibit practical failure modes when the unnormalised distribution of interest has isolated components -- they cannot discover isolated components or identify the correct mixing proportions between components. We demonstrate these findings using simple distribu"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2008.10087","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/2008.10087/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":"2008.10087","created_at":"2026-07-05T03:42:19.675828+00:00"},{"alias_kind":"arxiv_version","alias_value":"2008.10087v3","created_at":"2026-07-05T03:42:19.675828+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2008.10087","created_at":"2026-07-05T03:42:19.675828+00:00"},{"alias_kind":"pith_short_12","alias_value":"4HBCCS2OAGID","created_at":"2026-07-05T03:42:19.675828+00:00"},{"alias_kind":"pith_short_16","alias_value":"4HBCCS2OAGID2O2M","created_at":"2026-07-05T03:42:19.675828+00:00"},{"alias_kind":"pith_short_8","alias_value":"4HBCCS2O","created_at":"2026-07-05T03:42:19.675828+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":4,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.26850","citing_title":"Learning Energy-Based Models from Stochastic Interpolants using Spatiotemporal Differences","ref_index":77,"is_internal_anchor":false},{"citing_arxiv_id":"2605.22950","citing_title":"Diffusion-based Denoising Beats Vanilla Score Matching in Parameter Estimation: A Theoretical Explanation","ref_index":17,"is_internal_anchor":false},{"citing_arxiv_id":"2605.05118","citing_title":"On the Wasserstein Gradient Flow Interpretation of Drifting Models","ref_index":30,"is_internal_anchor":false},{"citing_arxiv_id":"2605.05118","citing_title":"On the Wasserstein Gradient Flow Interpretation of Drifting Models","ref_index":29,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/4HBCCS2OAGID2O2MYXTXRLRE7P","json":"https://pith.science/pith/4HBCCS2OAGID2O2MYXTXRLRE7P.json","graph_json":"https://pith.science/api/pith-number/4HBCCS2OAGID2O2MYXTXRLRE7P/graph.json","events_json":"https://pith.science/api/pith-number/4HBCCS2OAGID2O2MYXTXRLRE7P/events.json","paper":"https://pith.science/paper/4HBCCS2O"},"agent_actions":{"view_html":"https://pith.science/pith/4HBCCS2OAGID2O2MYXTXRLRE7P","download_json":"https://pith.science/pith/4HBCCS2OAGID2O2MYXTXRLRE7P.json","view_paper":"https://pith.science/paper/4HBCCS2O","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2008.10087&json=true","fetch_graph":"https://pith.science/api/pith-number/4HBCCS2OAGID2O2MYXTXRLRE7P/graph.json","fetch_events":"https://pith.science/api/pith-number/4HBCCS2OAGID2O2MYXTXRLRE7P/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/4HBCCS2OAGID2O2MYXTXRLRE7P/action/timestamp_anchor","attest_storage":"https://pith.science/pith/4HBCCS2OAGID2O2MYXTXRLRE7P/action/storage_attestation","attest_author":"https://pith.science/pith/4HBCCS2OAGID2O2MYXTXRLRE7P/action/author_attestation","sign_citation":"https://pith.science/pith/4HBCCS2OAGID2O2MYXTXRLRE7P/action/citation_signature","submit_replication":"https://pith.science/pith/4HBCCS2OAGID2O2MYXTXRLRE7P/action/replication_record"}},"created_at":"2026-07-05T03:42:19.675828+00:00","updated_at":"2026-07-05T03:42:19.675828+00:00"}