{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2015:YJQDLFIOLGRJGPBFWGXHUQKMCQ","short_pith_number":"pith:YJQDLFIO","schema_version":"1.0","canonical_sha256":"c26035950e59a2933c25b1ae7a414c140a4a994095c5bd5a2780baa2a19f91b2","source":{"kind":"arxiv","id":"1506.02158","version":6},"attestation_state":"computed","paper":{"title":"Bayesian Convolutional Neural Networks with Bernoulli Approximate Variational Inference","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"stat.ML","authors_text":"Yarin Gal, Zoubin Ghahramani","submitted_at":"2015-06-06T14:43:40Z","abstract_excerpt":"Convolutional neural networks (CNNs) work well on large datasets. But labelled data is hard to collect, and in some applications larger amounts of data are not available. The problem then is how to use CNNs with small data -- as CNNs overfit quickly. We present an efficient Bayesian CNN, offering better robustness to over-fitting on small data than traditional approaches. This is by placing a probability distribution over the CNN's kernels. We approximate our model's intractable posterior with Bernoulli variational distributions, requiring no additional model parameters.\n  On the theoretical s"},"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":"1506.02158","kind":"arxiv","version":6},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"stat.ML","submitted_at":"2015-06-06T14:43:40Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"6754fb552c4aa1a7f81025dc3eb30249e7047df0ddbfd2ce5042a0db86a354c2","abstract_canon_sha256":"25a028048580c260c9fdf16c2511c2edfed6107ea9d3577bc68f9e3100a5a588"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-18T01:22:48.413015Z","signature_b64":"AfbwBiakIJDYmw2KgRcU+ZfwYaiyHJ16H3x7Y04XrtQRZu9YBM+LSar935rQPWW7m7lMtPeFjkX6mgl3rdL4Cw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"c26035950e59a2933c25b1ae7a414c140a4a994095c5bd5a2780baa2a19f91b2","last_reissued_at":"2026-05-18T01:22:48.412471Z","signature_status":"signed_v1","first_computed_at":"2026-05-18T01:22:48.412471Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Bayesian Convolutional Neural Networks with Bernoulli Approximate Variational Inference","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"stat.ML","authors_text":"Yarin Gal, Zoubin Ghahramani","submitted_at":"2015-06-06T14:43:40Z","abstract_excerpt":"Convolutional neural networks (CNNs) work well on large datasets. But labelled data is hard to collect, and in some applications larger amounts of data are not available. The problem then is how to use CNNs with small data -- as CNNs overfit quickly. We present an efficient Bayesian CNN, offering better robustness to over-fitting on small data than traditional approaches. This is by placing a probability distribution over the CNN's kernels. We approximate our model's intractable posterior with Bernoulli variational distributions, requiring no additional model parameters.\n  On the theoretical s"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1506.02158","kind":"arxiv","version":6},"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":"1506.02158","created_at":"2026-05-18T01:22:48.412571+00:00"},{"alias_kind":"arxiv_version","alias_value":"1506.02158v6","created_at":"2026-05-18T01:22:48.412571+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1506.02158","created_at":"2026-05-18T01:22:48.412571+00:00"},{"alias_kind":"pith_short_12","alias_value":"YJQDLFIOLGRJ","created_at":"2026-05-18T12:29:50.041715+00:00"},{"alias_kind":"pith_short_16","alias_value":"YJQDLFIOLGRJGPBF","created_at":"2026-05-18T12:29:50.041715+00:00"},{"alias_kind":"pith_short_8","alias_value":"YJQDLFIO","created_at":"2026-05-18T12:29:50.041715+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":5,"internal_anchor_count":5,"sample":[{"citing_arxiv_id":"1906.10718","citing_title":"Active Learning Solution on Distributed Edge Computing","ref_index":23,"is_internal_anchor":true},{"citing_arxiv_id":"1907.05272","citing_title":"Introduction to Camera Pose Estimation with Deep Learning","ref_index":30,"is_internal_anchor":true},{"citing_arxiv_id":"1907.11202","citing_title":"Unsupervised Domain Adaptation via Calibrating Uncertainties","ref_index":4,"is_internal_anchor":true},{"citing_arxiv_id":"2412.10679","citing_title":"U-FaceBP: Uncertainty-aware Bayesian Ensemble Deep Learning for Face Video-based Blood Pressure Estimation","ref_index":44,"is_internal_anchor":true},{"citing_arxiv_id":"2605.14689","citing_title":"Are Candidate Models Really Needed for Active Learning?","ref_index":99,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/YJQDLFIOLGRJGPBFWGXHUQKMCQ","json":"https://pith.science/pith/YJQDLFIOLGRJGPBFWGXHUQKMCQ.json","graph_json":"https://pith.science/api/pith-number/YJQDLFIOLGRJGPBFWGXHUQKMCQ/graph.json","events_json":"https://pith.science/api/pith-number/YJQDLFIOLGRJGPBFWGXHUQKMCQ/events.json","paper":"https://pith.science/paper/YJQDLFIO"},"agent_actions":{"view_html":"https://pith.science/pith/YJQDLFIOLGRJGPBFWGXHUQKMCQ","download_json":"https://pith.science/pith/YJQDLFIOLGRJGPBFWGXHUQKMCQ.json","view_paper":"https://pith.science/paper/YJQDLFIO","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1506.02158&json=true","fetch_graph":"https://pith.science/api/pith-number/YJQDLFIOLGRJGPBFWGXHUQKMCQ/graph.json","fetch_events":"https://pith.science/api/pith-number/YJQDLFIOLGRJGPBFWGXHUQKMCQ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/YJQDLFIOLGRJGPBFWGXHUQKMCQ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/YJQDLFIOLGRJGPBFWGXHUQKMCQ/action/storage_attestation","attest_author":"https://pith.science/pith/YJQDLFIOLGRJGPBFWGXHUQKMCQ/action/author_attestation","sign_citation":"https://pith.science/pith/YJQDLFIOLGRJGPBFWGXHUQKMCQ/action/citation_signature","submit_replication":"https://pith.science/pith/YJQDLFIOLGRJGPBFWGXHUQKMCQ/action/replication_record"}},"created_at":"2026-05-18T01:22:48.412571+00:00","updated_at":"2026-05-18T01:22:48.412571+00:00"}