{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2019:HZQNR4F5ORDN5VITW7EMR4YH66","short_pith_number":"pith:HZQNR4F5","schema_version":"1.0","canonical_sha256":"3e60d8f0bd7446ded513b7c8c8f307f7b727e1845065f2fd97fdc194f7f5e34a","source":{"kind":"arxiv","id":"1906.03260","version":2},"attestation_state":"computed","paper":{"title":"Reliable training and estimation of variance networks","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"stat.ML","authors_text":"Martin J{\\o}rgensen, Nicki S. Detlefsen, S{\\o}ren Hauberg","submitted_at":"2019-06-04T10:14:58Z","abstract_excerpt":"We propose and investigate new complementary methodologies for estimating predictive variance networks in regression neural networks. We derive a locally aware mini-batching scheme that result in sparse robust gradients, and show how to make unbiased weight updates to a variance network. Further, we formulate a heuristic for robustly fitting both the mean and variance networks post hoc. Finally, we take inspiration from posterior Gaussian processes and propose a network architecture with similar extrapolation properties to Gaussian processes. The proposed methodologies are complementary, and i"},"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":"1906.03260","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"stat.ML","submitted_at":"2019-06-04T10:14:58Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"2d20ea640b2662fa37f6ad7260d52a61030be9488250bc3aa4b71acf57722acf","abstract_canon_sha256":"d529e13b92df4aaf5d1fd703b752352c1d7fa9feeb33ca4142fb18d5c90148dd"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T00:16:34.977629Z","signature_b64":"IWF1csJrSLAtopDGCZZRgOVUzVXVj1Fg0wOkBJGx5oeUl1+bllO0gkUKWIDnpmrRf+LYKdM4nG+kajKfwHD8Dw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"3e60d8f0bd7446ded513b7c8c8f307f7b727e1845065f2fd97fdc194f7f5e34a","last_reissued_at":"2026-07-05T00:16:34.977172Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T00:16:34.977172Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Reliable training and estimation of variance networks","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"stat.ML","authors_text":"Martin J{\\o}rgensen, Nicki S. Detlefsen, S{\\o}ren Hauberg","submitted_at":"2019-06-04T10:14:58Z","abstract_excerpt":"We propose and investigate new complementary methodologies for estimating predictive variance networks in regression neural networks. We derive a locally aware mini-batching scheme that result in sparse robust gradients, and show how to make unbiased weight updates to a variance network. Further, we formulate a heuristic for robustly fitting both the mean and variance networks post hoc. Finally, we take inspiration from posterior Gaussian processes and propose a network architecture with similar extrapolation properties to Gaussian processes. The proposed methodologies are complementary, and i"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1906.03260","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":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/1906.03260/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":"1906.03260","created_at":"2026-07-05T00:16:34.977228+00:00"},{"alias_kind":"arxiv_version","alias_value":"1906.03260v2","created_at":"2026-07-05T00:16:34.977228+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1906.03260","created_at":"2026-07-05T00:16:34.977228+00:00"},{"alias_kind":"pith_short_12","alias_value":"HZQNR4F5ORDN","created_at":"2026-07-05T00:16:34.977228+00:00"},{"alias_kind":"pith_short_16","alias_value":"HZQNR4F5ORDN5VIT","created_at":"2026-07-05T00:16:34.977228+00:00"},{"alias_kind":"pith_short_8","alias_value":"HZQNR4F5","created_at":"2026-07-05T00:16:34.977228+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2509.00155","citing_title":"Amplitude Uncertainties Everywhere All at Once","ref_index":67,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/HZQNR4F5ORDN5VITW7EMR4YH66","json":"https://pith.science/pith/HZQNR4F5ORDN5VITW7EMR4YH66.json","graph_json":"https://pith.science/api/pith-number/HZQNR4F5ORDN5VITW7EMR4YH66/graph.json","events_json":"https://pith.science/api/pith-number/HZQNR4F5ORDN5VITW7EMR4YH66/events.json","paper":"https://pith.science/paper/HZQNR4F5"},"agent_actions":{"view_html":"https://pith.science/pith/HZQNR4F5ORDN5VITW7EMR4YH66","download_json":"https://pith.science/pith/HZQNR4F5ORDN5VITW7EMR4YH66.json","view_paper":"https://pith.science/paper/HZQNR4F5","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1906.03260&json=true","fetch_graph":"https://pith.science/api/pith-number/HZQNR4F5ORDN5VITW7EMR4YH66/graph.json","fetch_events":"https://pith.science/api/pith-number/HZQNR4F5ORDN5VITW7EMR4YH66/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/HZQNR4F5ORDN5VITW7EMR4YH66/action/timestamp_anchor","attest_storage":"https://pith.science/pith/HZQNR4F5ORDN5VITW7EMR4YH66/action/storage_attestation","attest_author":"https://pith.science/pith/HZQNR4F5ORDN5VITW7EMR4YH66/action/author_attestation","sign_citation":"https://pith.science/pith/HZQNR4F5ORDN5VITW7EMR4YH66/action/citation_signature","submit_replication":"https://pith.science/pith/HZQNR4F5ORDN5VITW7EMR4YH66/action/replication_record"}},"created_at":"2026-07-05T00:16:34.977228+00:00","updated_at":"2026-07-05T00:16:34.977228+00:00"}