{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2019:CWITVAH7CJN2Q3BU7OM3WHJFGF","short_pith_number":"pith:CWITVAH7","schema_version":"1.0","canonical_sha256":"15913a80ff125ba86c34fb99bb1d25316bb4b34c8ffc1d9ffe3f8df316bdeac4","source":{"kind":"arxiv","id":"1906.06914","version":1},"attestation_state":"computed","paper":{"title":"Variational Inference with Numerical Derivatives: variance reduction through coupling","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"stat.CO","authors_text":"Alexander Immer, Guillaume P. Dehaene","submitted_at":"2019-06-17T09:30:29Z","abstract_excerpt":"The Black Box Variational Inference (Ranganath et al. (2014)) algorithm provides a universal method for Variational Inference, but taking advantage of special properties of the approximation family or of the target can improve the convergence speed significantly. For example, if the approximation family is a transformation family, such as a Gaussian, then switching to the reparameterization gradient (Kingma and Welling (2014)) often yields a major reduction in gradient variance. Ultimately, reducing the variance can reduce the computational cost and yield better approximations.\n  We present 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":"1906.06914","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"stat.CO","submitted_at":"2019-06-17T09:30:29Z","cross_cats_sorted":["stat.ML"],"title_canon_sha256":"197bdaebe63e09c1ff01e278d5f3bd01bad2942d32273b272f9cbf8b30912f49","abstract_canon_sha256":"6ff2c709e4a0a215b595cd4d280c06f8ab4ad74e8ed5c3ea3288151f988b7ae0"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-17T23:43:12.338070Z","signature_b64":"D7OeipJ9sooq4Ajgm3bANcoZodFMLF3ugdrZOEB8ukLSkeV3l/upOQcK++G5zdBNPzRFkrv//bt5tuHwr4WtCg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"15913a80ff125ba86c34fb99bb1d25316bb4b34c8ffc1d9ffe3f8df316bdeac4","last_reissued_at":"2026-05-17T23:43:12.337595Z","signature_status":"signed_v1","first_computed_at":"2026-05-17T23:43:12.337595Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Variational Inference with Numerical Derivatives: variance reduction through coupling","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"stat.CO","authors_text":"Alexander Immer, Guillaume P. Dehaene","submitted_at":"2019-06-17T09:30:29Z","abstract_excerpt":"The Black Box Variational Inference (Ranganath et al. (2014)) algorithm provides a universal method for Variational Inference, but taking advantage of special properties of the approximation family or of the target can improve the convergence speed significantly. For example, if the approximation family is a transformation family, such as a Gaussian, then switching to the reparameterization gradient (Kingma and Welling (2014)) often yields a major reduction in gradient variance. Ultimately, reducing the variance can reduce the computational cost and yield better approximations.\n  We present a "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1906.06914","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":""},"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":"1906.06914","created_at":"2026-05-17T23:43:12.337672+00:00"},{"alias_kind":"arxiv_version","alias_value":"1906.06914v1","created_at":"2026-05-17T23:43:12.337672+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1906.06914","created_at":"2026-05-17T23:43:12.337672+00:00"},{"alias_kind":"pith_short_12","alias_value":"CWITVAH7CJN2","created_at":"2026-05-18T12:33:15.570797+00:00"},{"alias_kind":"pith_short_16","alias_value":"CWITVAH7CJN2Q3BU","created_at":"2026-05-18T12:33:15.570797+00:00"},{"alias_kind":"pith_short_8","alias_value":"CWITVAH7","created_at":"2026-05-18T12:33:15.570797+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/CWITVAH7CJN2Q3BU7OM3WHJFGF","json":"https://pith.science/pith/CWITVAH7CJN2Q3BU7OM3WHJFGF.json","graph_json":"https://pith.science/api/pith-number/CWITVAH7CJN2Q3BU7OM3WHJFGF/graph.json","events_json":"https://pith.science/api/pith-number/CWITVAH7CJN2Q3BU7OM3WHJFGF/events.json","paper":"https://pith.science/paper/CWITVAH7"},"agent_actions":{"view_html":"https://pith.science/pith/CWITVAH7CJN2Q3BU7OM3WHJFGF","download_json":"https://pith.science/pith/CWITVAH7CJN2Q3BU7OM3WHJFGF.json","view_paper":"https://pith.science/paper/CWITVAH7","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1906.06914&json=true","fetch_graph":"https://pith.science/api/pith-number/CWITVAH7CJN2Q3BU7OM3WHJFGF/graph.json","fetch_events":"https://pith.science/api/pith-number/CWITVAH7CJN2Q3BU7OM3WHJFGF/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/CWITVAH7CJN2Q3BU7OM3WHJFGF/action/timestamp_anchor","attest_storage":"https://pith.science/pith/CWITVAH7CJN2Q3BU7OM3WHJFGF/action/storage_attestation","attest_author":"https://pith.science/pith/CWITVAH7CJN2Q3BU7OM3WHJFGF/action/author_attestation","sign_citation":"https://pith.science/pith/CWITVAH7CJN2Q3BU7OM3WHJFGF/action/citation_signature","submit_replication":"https://pith.science/pith/CWITVAH7CJN2Q3BU7OM3WHJFGF/action/replication_record"}},"created_at":"2026-05-17T23:43:12.337672+00:00","updated_at":"2026-05-17T23:43:12.337672+00:00"}