{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2015:VGKJDVVOYMQHSHPXQUFGHTSFFQ","short_pith_number":"pith:VGKJDVVO","schema_version":"1.0","canonical_sha256":"a99491d6aec320791df7850a63ce452c08a6949bf4b9ca6ffb5dd7d3dc68b24c","source":{"kind":"arxiv","id":"1505.05424","version":2},"attestation_state":"computed","paper":{"title":"Weight Uncertainty in Neural Networks","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"stat.ML","authors_text":"Charles Blundell, Daan Wierstra, Julien Cornebise, Koray Kavukcuoglu","submitted_at":"2015-05-20T15:39:48Z","abstract_excerpt":"We introduce a new, efficient, principled and backpropagation-compatible algorithm for learning a probability distribution on the weights of a neural network, called Bayes by Backprop. It regularises the weights by minimising a compression cost, known as the variational free energy or the expected lower bound on the marginal likelihood. We show that this principled kind of regularisation yields comparable performance to dropout on MNIST classification. We then demonstrate how the learnt uncertainty in the weights can be used to improve generalisation in non-linear regression problems, and how "},"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":"1505.05424","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"stat.ML","submitted_at":"2015-05-20T15:39:48Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"eaf2e08aee1640ec9968444211934c808c55c7b67672e5c20e6c1872fac3fd8d","abstract_canon_sha256":"8f2732c96775298310c969e934a178f3c47a78dbbe819e9c8e1089d2d13cbe56"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-18T02:03:54.495109Z","signature_b64":"qrI83ai11yKlWpGW4HFa754H8VuoKti2jk7U+Lo0GbZ8jBro0XZywK/930QZA5vuAH5K5nqewlAjvPZNR9dSCg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"a99491d6aec320791df7850a63ce452c08a6949bf4b9ca6ffb5dd7d3dc68b24c","last_reissued_at":"2026-05-18T02:03:54.494296Z","signature_status":"signed_v1","first_computed_at":"2026-05-18T02:03:54.494296Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Weight Uncertainty in Neural Networks","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"stat.ML","authors_text":"Charles Blundell, Daan Wierstra, Julien Cornebise, Koray Kavukcuoglu","submitted_at":"2015-05-20T15:39:48Z","abstract_excerpt":"We introduce a new, efficient, principled and backpropagation-compatible algorithm for learning a probability distribution on the weights of a neural network, called Bayes by Backprop. It regularises the weights by minimising a compression cost, known as the variational free energy or the expected lower bound on the marginal likelihood. We show that this principled kind of regularisation yields comparable performance to dropout on MNIST classification. We then demonstrate how the learnt uncertainty in the weights can be used to improve generalisation in non-linear regression problems, and how "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1505.05424","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":"1505.05424","created_at":"2026-05-18T02:03:54.494444+00:00"},{"alias_kind":"arxiv_version","alias_value":"1505.05424v2","created_at":"2026-05-18T02:03:54.494444+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1505.05424","created_at":"2026-05-18T02:03:54.494444+00:00"},{"alias_kind":"pith_short_12","alias_value":"VGKJDVVOYMQH","created_at":"2026-05-18T12:29:44.643036+00:00"},{"alias_kind":"pith_short_16","alias_value":"VGKJDVVOYMQHSHPX","created_at":"2026-05-18T12:29:44.643036+00:00"},{"alias_kind":"pith_short_8","alias_value":"VGKJDVVO","created_at":"2026-05-18T12:29:44.643036+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":7,"internal_anchor_count":5,"sample":[{"citing_arxiv_id":"1906.09686","citing_title":"Quality of Uncertainty Quantification for Bayesian Neural Network Inference","ref_index":1,"is_internal_anchor":true},{"citing_arxiv_id":"1906.11909","citing_title":"Comparing Semi-Parametric Model Learning Algorithms for Dynamic Model Estimation in Robotics","ref_index":14,"is_internal_anchor":true},{"citing_arxiv_id":"1907.11202","citing_title":"Unsupervised Domain Adaptation via Calibrating Uncertainties","ref_index":1,"is_internal_anchor":true},{"citing_arxiv_id":"2006.12024","citing_title":"Bayesian Neural Networks: An Introduction and Survey","ref_index":76,"is_internal_anchor":true},{"citing_arxiv_id":"2402.14532","citing_title":"A Framework for Variational Inference of Lightweight Bayesian Neural Networks with Heteroscedastic Uncertainties","ref_index":17,"is_internal_anchor":true},{"citing_arxiv_id":"2604.14206","citing_title":"Portfolio Optimization Proxies under Label Scarcity and Regime Shifts via Bayesian and Deterministic Students under Semi-Supervised Sandwich Training","ref_index":9,"is_internal_anchor":false},{"citing_arxiv_id":"1606.06565","citing_title":"Concrete Problems in AI Safety","ref_index":26,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/VGKJDVVOYMQHSHPXQUFGHTSFFQ","json":"https://pith.science/pith/VGKJDVVOYMQHSHPXQUFGHTSFFQ.json","graph_json":"https://pith.science/api/pith-number/VGKJDVVOYMQHSHPXQUFGHTSFFQ/graph.json","events_json":"https://pith.science/api/pith-number/VGKJDVVOYMQHSHPXQUFGHTSFFQ/events.json","paper":"https://pith.science/paper/VGKJDVVO"},"agent_actions":{"view_html":"https://pith.science/pith/VGKJDVVOYMQHSHPXQUFGHTSFFQ","download_json":"https://pith.science/pith/VGKJDVVOYMQHSHPXQUFGHTSFFQ.json","view_paper":"https://pith.science/paper/VGKJDVVO","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1505.05424&json=true","fetch_graph":"https://pith.science/api/pith-number/VGKJDVVOYMQHSHPXQUFGHTSFFQ/graph.json","fetch_events":"https://pith.science/api/pith-number/VGKJDVVOYMQHSHPXQUFGHTSFFQ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/VGKJDVVOYMQHSHPXQUFGHTSFFQ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/VGKJDVVOYMQHSHPXQUFGHTSFFQ/action/storage_attestation","attest_author":"https://pith.science/pith/VGKJDVVOYMQHSHPXQUFGHTSFFQ/action/author_attestation","sign_citation":"https://pith.science/pith/VGKJDVVOYMQHSHPXQUFGHTSFFQ/action/citation_signature","submit_replication":"https://pith.science/pith/VGKJDVVOYMQHSHPXQUFGHTSFFQ/action/replication_record"}},"created_at":"2026-05-18T02:03:54.494444+00:00","updated_at":"2026-05-18T02:03:54.494444+00:00"}