{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2018:4637WB3OB5W2U4EX43MNIIWWEH","short_pith_number":"pith:4637WB3O","schema_version":"1.0","canonical_sha256":"e7b7fb076e0f6daa7097e6d8d422d621ec9ffb30810ab87e949e68dc4ba745b7","source":{"kind":"arxiv","id":"1811.08357","version":4},"attestation_state":"computed","paper":{"title":"Learning deep kernels for exponential family densities","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG","stat.ME"],"primary_cat":"stat.ML","authors_text":"Arthur Gretton, Danica J. Sutherland, Heiko Strathmann, Li Wenliang","submitted_at":"2018-11-20T16:40:45Z","abstract_excerpt":"The kernel exponential family is a rich class of distributions, which can be fit efficiently and with statistical guarantees by score matching. Being required to choose a priori a simple kernel such as the Gaussian, however, limits its practical applicability. We provide a scheme for learning a kernel parameterized by a deep network, which can find complex location-dependent local features of the data geometry. This gives a very rich class of density models, capable of fitting complex structures on moderate-dimensional problems. Compared to deep density models fit via maximum likelihood, our 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":"1811.08357","kind":"arxiv","version":4},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"stat.ML","submitted_at":"2018-11-20T16:40:45Z","cross_cats_sorted":["cs.LG","stat.ME"],"title_canon_sha256":"b8705b9f1535ea027e5c13feed6bd33dbf6807f989fad74186fca118f487f9b1","abstract_canon_sha256":"886873b74800b666bdbd05e4e9125d3533e048175f5c19325ad367cada891bea"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T02:06:47.007024Z","signature_b64":"EegjOUPvvcc+9KANON5pLizAguT1O0jrv1GZAuV7fPN4P9FH0/ROYL2m/e3TrZNbxjHctFbgDi7z/NPE1jwnDw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"e7b7fb076e0f6daa7097e6d8d422d621ec9ffb30810ab87e949e68dc4ba745b7","last_reissued_at":"2026-07-05T02:06:47.006567Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T02:06:47.006567Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Learning deep kernels for exponential family densities","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG","stat.ME"],"primary_cat":"stat.ML","authors_text":"Arthur Gretton, Danica J. Sutherland, Heiko Strathmann, Li Wenliang","submitted_at":"2018-11-20T16:40:45Z","abstract_excerpt":"The kernel exponential family is a rich class of distributions, which can be fit efficiently and with statistical guarantees by score matching. Being required to choose a priori a simple kernel such as the Gaussian, however, limits its practical applicability. We provide a scheme for learning a kernel parameterized by a deep network, which can find complex location-dependent local features of the data geometry. This gives a very rich class of density models, capable of fitting complex structures on moderate-dimensional problems. Compared to deep density models fit via maximum likelihood, our a"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1811.08357","kind":"arxiv","version":4},"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/1811.08357/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":"1811.08357","created_at":"2026-07-05T02:06:47.006625+00:00"},{"alias_kind":"arxiv_version","alias_value":"1811.08357v4","created_at":"2026-07-05T02:06:47.006625+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1811.08357","created_at":"2026-07-05T02:06:47.006625+00:00"},{"alias_kind":"pith_short_12","alias_value":"4637WB3OB5W2","created_at":"2026-07-05T02:06:47.006625+00:00"},{"alias_kind":"pith_short_16","alias_value":"4637WB3OB5W2U4EX","created_at":"2026-07-05T02:06:47.006625+00:00"},{"alias_kind":"pith_short_8","alias_value":"4637WB3O","created_at":"2026-07-05T02:06:47.006625+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2506.09258","citing_title":"CFMI: Flow Matching for Missing Data Imputation","ref_index":67,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/4637WB3OB5W2U4EX43MNIIWWEH","json":"https://pith.science/pith/4637WB3OB5W2U4EX43MNIIWWEH.json","graph_json":"https://pith.science/api/pith-number/4637WB3OB5W2U4EX43MNIIWWEH/graph.json","events_json":"https://pith.science/api/pith-number/4637WB3OB5W2U4EX43MNIIWWEH/events.json","paper":"https://pith.science/paper/4637WB3O"},"agent_actions":{"view_html":"https://pith.science/pith/4637WB3OB5W2U4EX43MNIIWWEH","download_json":"https://pith.science/pith/4637WB3OB5W2U4EX43MNIIWWEH.json","view_paper":"https://pith.science/paper/4637WB3O","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1811.08357&json=true","fetch_graph":"https://pith.science/api/pith-number/4637WB3OB5W2U4EX43MNIIWWEH/graph.json","fetch_events":"https://pith.science/api/pith-number/4637WB3OB5W2U4EX43MNIIWWEH/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/4637WB3OB5W2U4EX43MNIIWWEH/action/timestamp_anchor","attest_storage":"https://pith.science/pith/4637WB3OB5W2U4EX43MNIIWWEH/action/storage_attestation","attest_author":"https://pith.science/pith/4637WB3OB5W2U4EX43MNIIWWEH/action/author_attestation","sign_citation":"https://pith.science/pith/4637WB3OB5W2U4EX43MNIIWWEH/action/citation_signature","submit_replication":"https://pith.science/pith/4637WB3OB5W2U4EX43MNIIWWEH/action/replication_record"}},"created_at":"2026-07-05T02:06:47.006625+00:00","updated_at":"2026-07-05T02:06:47.006625+00:00"}