{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2017:YRDA3RB7O6WLKTQ37QCMJUJJAO","short_pith_number":"pith:YRDA3RB7","schema_version":"1.0","canonical_sha256":"c4460dc43f77acb54e1bfc04c4d1290397d48a57a6dd9a3bb02c29a369202b18","source":{"kind":"arxiv","id":"1705.10388","version":1},"attestation_state":"computed","paper":{"title":"Model Selection in Bayesian Neural Networks via Horseshoe Priors","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"stat.ML","authors_text":"Finale Doshi-Velez, Soumya Ghosh","submitted_at":"2017-05-29T20:35:42Z","abstract_excerpt":"Bayesian Neural Networks (BNNs) have recently received increasing attention for their ability to provide well-calibrated posterior uncertainties. However, model selection---even choosing the number of nodes---remains an open question. In this work, we apply a horseshoe prior over node pre-activations of a Bayesian neural network, which effectively turns off nodes that do not help explain the data. We demonstrate that our prior prevents the BNN from under-fitting even when the number of nodes required is grossly over-estimated. Moreover, this model selection over the number of nodes doesn't com"},"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":"1705.10388","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"stat.ML","submitted_at":"2017-05-29T20:35:42Z","cross_cats_sorted":[],"title_canon_sha256":"ba9e9e7f241e710a20ca0c527adcacb15fe03e0a92dd281544d3033f2bf7b496","abstract_canon_sha256":"82e24c152448d4bea6ff42c15c0a2d87cf9d913feb2c7d49c6bf179364a3968a"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-18T00:43:31.187023Z","signature_b64":"o02VnzSua/tVOC2VcsxUpdBc9kIaWxJX/Mti9BrCC6p4VzvGjgoTaUMIdaiGqEL6u3Nk+Buldosxr9okEkQXBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"c4460dc43f77acb54e1bfc04c4d1290397d48a57a6dd9a3bb02c29a369202b18","last_reissued_at":"2026-05-18T00:43:31.186481Z","signature_status":"signed_v1","first_computed_at":"2026-05-18T00:43:31.186481Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Model Selection in Bayesian Neural Networks via Horseshoe Priors","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"stat.ML","authors_text":"Finale Doshi-Velez, Soumya Ghosh","submitted_at":"2017-05-29T20:35:42Z","abstract_excerpt":"Bayesian Neural Networks (BNNs) have recently received increasing attention for their ability to provide well-calibrated posterior uncertainties. However, model selection---even choosing the number of nodes---remains an open question. In this work, we apply a horseshoe prior over node pre-activations of a Bayesian neural network, which effectively turns off nodes that do not help explain the data. We demonstrate that our prior prevents the BNN from under-fitting even when the number of nodes required is grossly over-estimated. Moreover, this model selection over the number of nodes doesn't com"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1705.10388","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":"1705.10388","created_at":"2026-05-18T00:43:31.186567+00:00"},{"alias_kind":"arxiv_version","alias_value":"1705.10388v1","created_at":"2026-05-18T00:43:31.186567+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1705.10388","created_at":"2026-05-18T00:43:31.186567+00:00"},{"alias_kind":"pith_short_12","alias_value":"YRDA3RB7O6WL","created_at":"2026-05-18T12:31:56.362134+00:00"},{"alias_kind":"pith_short_16","alias_value":"YRDA3RB7O6WLKTQ3","created_at":"2026-05-18T12:31:56.362134+00:00"},{"alias_kind":"pith_short_8","alias_value":"YRDA3RB7","created_at":"2026-05-18T12:31:56.362134+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.05493","citing_title":"A renormalization-group inspired lattice-based framework for piecewise generalized linear models","ref_index":221,"is_internal_anchor":false},{"citing_arxiv_id":"2605.05396","citing_title":"Bayesian Region Selection and Prediction in Poisson Regression with Spatially Dependent Global-Local Shrinkage Prior","ref_index":79,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/YRDA3RB7O6WLKTQ37QCMJUJJAO","json":"https://pith.science/pith/YRDA3RB7O6WLKTQ37QCMJUJJAO.json","graph_json":"https://pith.science/api/pith-number/YRDA3RB7O6WLKTQ37QCMJUJJAO/graph.json","events_json":"https://pith.science/api/pith-number/YRDA3RB7O6WLKTQ37QCMJUJJAO/events.json","paper":"https://pith.science/paper/YRDA3RB7"},"agent_actions":{"view_html":"https://pith.science/pith/YRDA3RB7O6WLKTQ37QCMJUJJAO","download_json":"https://pith.science/pith/YRDA3RB7O6WLKTQ37QCMJUJJAO.json","view_paper":"https://pith.science/paper/YRDA3RB7","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1705.10388&json=true","fetch_graph":"https://pith.science/api/pith-number/YRDA3RB7O6WLKTQ37QCMJUJJAO/graph.json","fetch_events":"https://pith.science/api/pith-number/YRDA3RB7O6WLKTQ37QCMJUJJAO/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/YRDA3RB7O6WLKTQ37QCMJUJJAO/action/timestamp_anchor","attest_storage":"https://pith.science/pith/YRDA3RB7O6WLKTQ37QCMJUJJAO/action/storage_attestation","attest_author":"https://pith.science/pith/YRDA3RB7O6WLKTQ37QCMJUJJAO/action/author_attestation","sign_citation":"https://pith.science/pith/YRDA3RB7O6WLKTQ37QCMJUJJAO/action/citation_signature","submit_replication":"https://pith.science/pith/YRDA3RB7O6WLKTQ37QCMJUJJAO/action/replication_record"}},"created_at":"2026-05-18T00:43:31.186567+00:00","updated_at":"2026-05-18T00:43:31.186567+00:00"}