{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:NTVPFVF2N3YYMKO6ARLBCW3KG4","short_pith_number":"pith:NTVPFVF2","schema_version":"1.0","canonical_sha256":"6ceaf2d4ba6ef18629de0456115b6a371defbe1b08408b79250f313fb3dc54ec","source":{"kind":"arxiv","id":"2110.01593","version":8},"attestation_state":"computed","paper":{"title":"Generalized Kernel Thinning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG","math.ST","stat.ME","stat.TH"],"primary_cat":"stat.ML","authors_text":"Lester Mackey, Raaz Dwivedi","submitted_at":"2021-10-04T17:41:53Z","abstract_excerpt":"The kernel thinning (KT) algorithm of Dwivedi and Mackey (2021) compresses a probability distribution more effectively than independent sampling by targeting a reproducing kernel Hilbert space (RKHS) and leveraging a less smooth square-root kernel. Here we provide four improvements. First, we show that KT applied directly to the target RKHS yields tighter, dimension-free guarantees for any kernel, any distribution, and any fixed function in the RKHS. Second, we show that, for analytic kernels like Gaussian, inverse multiquadric, and sinc, target KT admits maximum mean discrepancy (MMD) guarant"},"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":"2110.01593","kind":"arxiv","version":8},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"stat.ML","submitted_at":"2021-10-04T17:41:53Z","cross_cats_sorted":["cs.LG","math.ST","stat.ME","stat.TH"],"title_canon_sha256":"094402e9d5f9cdbed7b81414ec4c18ac133d9654f82aab234178f6daf131b410","abstract_canon_sha256":"007fbdc60080682fca4c6c4e71658671e900c4a830294b18b0e79132ca16cc78"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:02:38.885101Z","signature_b64":"LhvSrWGewK24C52WBKRA+g2jmKJYe+Ef1uLI0Su5u77sVCM9WRB4G16DlVVsMI6xU0jdtOSX6qHWfYF2J7RtDg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"6ceaf2d4ba6ef18629de0456115b6a371defbe1b08408b79250f313fb3dc54ec","last_reissued_at":"2026-07-05T10:02:38.884664Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:02:38.884664Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Generalized Kernel Thinning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG","math.ST","stat.ME","stat.TH"],"primary_cat":"stat.ML","authors_text":"Lester Mackey, Raaz Dwivedi","submitted_at":"2021-10-04T17:41:53Z","abstract_excerpt":"The kernel thinning (KT) algorithm of Dwivedi and Mackey (2021) compresses a probability distribution more effectively than independent sampling by targeting a reproducing kernel Hilbert space (RKHS) and leveraging a less smooth square-root kernel. Here we provide four improvements. First, we show that KT applied directly to the target RKHS yields tighter, dimension-free guarantees for any kernel, any distribution, and any fixed function in the RKHS. Second, we show that, for analytic kernels like Gaussian, inverse multiquadric, and sinc, target KT admits maximum mean discrepancy (MMD) guarant"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2110.01593","kind":"arxiv","version":8},"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/2110.01593/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":"2110.01593","created_at":"2026-07-05T10:02:38.884732+00:00"},{"alias_kind":"arxiv_version","alias_value":"2110.01593v8","created_at":"2026-07-05T10:02:38.884732+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2110.01593","created_at":"2026-07-05T10:02:38.884732+00:00"},{"alias_kind":"pith_short_12","alias_value":"NTVPFVF2N3YY","created_at":"2026-07-05T10:02:38.884732+00:00"},{"alias_kind":"pith_short_16","alias_value":"NTVPFVF2N3YYMKO6","created_at":"2026-07-05T10:02:38.884732+00:00"},{"alias_kind":"pith_short_8","alias_value":"NTVPFVF2","created_at":"2026-07-05T10:02:38.884732+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2502.10600","citing_title":"Weighted quantization using MMD: From mean field to mean shift via gradient flows","ref_index":30,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/NTVPFVF2N3YYMKO6ARLBCW3KG4","json":"https://pith.science/pith/NTVPFVF2N3YYMKO6ARLBCW3KG4.json","graph_json":"https://pith.science/api/pith-number/NTVPFVF2N3YYMKO6ARLBCW3KG4/graph.json","events_json":"https://pith.science/api/pith-number/NTVPFVF2N3YYMKO6ARLBCW3KG4/events.json","paper":"https://pith.science/paper/NTVPFVF2"},"agent_actions":{"view_html":"https://pith.science/pith/NTVPFVF2N3YYMKO6ARLBCW3KG4","download_json":"https://pith.science/pith/NTVPFVF2N3YYMKO6ARLBCW3KG4.json","view_paper":"https://pith.science/paper/NTVPFVF2","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2110.01593&json=true","fetch_graph":"https://pith.science/api/pith-number/NTVPFVF2N3YYMKO6ARLBCW3KG4/graph.json","fetch_events":"https://pith.science/api/pith-number/NTVPFVF2N3YYMKO6ARLBCW3KG4/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/NTVPFVF2N3YYMKO6ARLBCW3KG4/action/timestamp_anchor","attest_storage":"https://pith.science/pith/NTVPFVF2N3YYMKO6ARLBCW3KG4/action/storage_attestation","attest_author":"https://pith.science/pith/NTVPFVF2N3YYMKO6ARLBCW3KG4/action/author_attestation","sign_citation":"https://pith.science/pith/NTVPFVF2N3YYMKO6ARLBCW3KG4/action/citation_signature","submit_replication":"https://pith.science/pith/NTVPFVF2N3YYMKO6ARLBCW3KG4/action/replication_record"}},"created_at":"2026-07-05T10:02:38.884732+00:00","updated_at":"2026-07-05T10:02:38.884732+00:00"}