{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2015:7URLFG4GZZM4HNOWHXDV43XABM","short_pith_number":"pith:7URLFG4G","schema_version":"1.0","canonical_sha256":"fd22b29b86ce59c3b5d63dc75e6ee00b1988d4c7d940fe73ebbeba0c98470a59","source":{"kind":"arxiv","id":"1506.02557","version":2},"attestation_state":"computed","paper":{"title":"Variational Dropout and the Local Reparameterization Trick","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG","stat.CO"],"primary_cat":"stat.ML","authors_text":"Diederik P. Kingma, Max Welling, Tim Salimans","submitted_at":"2015-06-08T15:37:56Z","abstract_excerpt":"We investigate a local reparameterizaton technique for greatly reducing the variance of stochastic gradients for variational Bayesian inference (SGVB) of a posterior over model parameters, while retaining parallelizability. This local reparameterization translates uncertainty about global parameters into local noise that is independent across datapoints in the minibatch. Such parameterizations can be trivially parallelized and have variance that is inversely proportional to the minibatch size, generally leading to much faster convergence. Additionally, we explore a connection with dropout: Gau"},"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":"1506.02557","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"stat.ML","submitted_at":"2015-06-08T15:37:56Z","cross_cats_sorted":["cs.LG","stat.CO"],"title_canon_sha256":"aeb1b8e9fd51efe690311011030c2a1de90f503222e3a54ed482c463986a5469","abstract_canon_sha256":"5ad34b7255c8a8af185aad436b63d2034f657f4b3b7521e5c750d62b8bda46c9"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-18T01:24:03.639958Z","signature_b64":"1cGcRhU+QyajVECeXA9zKzf8nmnpI8ZrGqCT/iOTFY4tx7D2Gl7f1jzcZ5N5tUDCBJpKodNYxQ3/uZzdD+1KCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"fd22b29b86ce59c3b5d63dc75e6ee00b1988d4c7d940fe73ebbeba0c98470a59","last_reissued_at":"2026-05-18T01:24:03.639335Z","signature_status":"signed_v1","first_computed_at":"2026-05-18T01:24:03.639335Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Variational Dropout and the Local Reparameterization Trick","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG","stat.CO"],"primary_cat":"stat.ML","authors_text":"Diederik P. Kingma, Max Welling, Tim Salimans","submitted_at":"2015-06-08T15:37:56Z","abstract_excerpt":"We investigate a local reparameterizaton technique for greatly reducing the variance of stochastic gradients for variational Bayesian inference (SGVB) of a posterior over model parameters, while retaining parallelizability. This local reparameterization translates uncertainty about global parameters into local noise that is independent across datapoints in the minibatch. Such parameterizations can be trivially parallelized and have variance that is inversely proportional to the minibatch size, generally leading to much faster convergence. Additionally, we explore a connection with dropout: Gau"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1506.02557","kind":"arxiv","version":2},"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":"1506.02557","created_at":"2026-05-18T01:24:03.639414+00:00"},{"alias_kind":"arxiv_version","alias_value":"1506.02557v2","created_at":"2026-05-18T01:24:03.639414+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1506.02557","created_at":"2026-05-18T01:24:03.639414+00:00"},{"alias_kind":"pith_short_12","alias_value":"7URLFG4GZZM4","created_at":"2026-05-18T12:29:10.953037+00:00"},{"alias_kind":"pith_short_16","alias_value":"7URLFG4GZZM4HNOW","created_at":"2026-05-18T12:29:10.953037+00:00"},{"alias_kind":"pith_short_8","alias_value":"7URLFG4G","created_at":"2026-05-18T12:29:10.953037+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":3,"internal_anchor_count":2,"sample":[{"citing_arxiv_id":"2502.02345","citing_title":"Low Rank Based Subspace Inference for the Laplace Approximation of Bayesian Neural Networks","ref_index":7,"is_internal_anchor":true},{"citing_arxiv_id":"2502.05564","citing_title":"TabICL: A Tabular Foundation Model for In-Context Learning on Large Data","ref_index":270,"is_internal_anchor":true},{"citing_arxiv_id":"2604.06032","citing_title":"Ensemble-Based Dirichlet Modeling for Predictive Uncertainty and Selective Classification","ref_index":4,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/7URLFG4GZZM4HNOWHXDV43XABM","json":"https://pith.science/pith/7URLFG4GZZM4HNOWHXDV43XABM.json","graph_json":"https://pith.science/api/pith-number/7URLFG4GZZM4HNOWHXDV43XABM/graph.json","events_json":"https://pith.science/api/pith-number/7URLFG4GZZM4HNOWHXDV43XABM/events.json","paper":"https://pith.science/paper/7URLFG4G"},"agent_actions":{"view_html":"https://pith.science/pith/7URLFG4GZZM4HNOWHXDV43XABM","download_json":"https://pith.science/pith/7URLFG4GZZM4HNOWHXDV43XABM.json","view_paper":"https://pith.science/paper/7URLFG4G","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1506.02557&json=true","fetch_graph":"https://pith.science/api/pith-number/7URLFG4GZZM4HNOWHXDV43XABM/graph.json","fetch_events":"https://pith.science/api/pith-number/7URLFG4GZZM4HNOWHXDV43XABM/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/7URLFG4GZZM4HNOWHXDV43XABM/action/timestamp_anchor","attest_storage":"https://pith.science/pith/7URLFG4GZZM4HNOWHXDV43XABM/action/storage_attestation","attest_author":"https://pith.science/pith/7URLFG4GZZM4HNOWHXDV43XABM/action/author_attestation","sign_citation":"https://pith.science/pith/7URLFG4GZZM4HNOWHXDV43XABM/action/citation_signature","submit_replication":"https://pith.science/pith/7URLFG4GZZM4HNOWHXDV43XABM/action/replication_record"}},"created_at":"2026-05-18T01:24:03.639414+00:00","updated_at":"2026-05-18T01:24:03.639414+00:00"}