{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2020:YV76XCEQH3R4N74XCTCGZHPD5T","short_pith_number":"pith:YV76XCEQ","schema_version":"1.0","canonical_sha256":"c57feb88903ee3c6ff9714c46c9de3ecd801bdaeb2cc89ad3541689bddd37a2b","source":{"kind":"arxiv","id":"2011.00716","version":5},"attestation_state":"computed","paper":{"title":"PAC Confidence Predictions for Deep Neural Network Classifiers","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Insup Lee, Osbert Bastani, Sangdon Park, Shuo Li","submitted_at":"2020-11-02T04:09:17Z","abstract_excerpt":"A key challenge for deploying deep neural networks (DNNs) in safety critical settings is the need to provide rigorous ways to quantify their uncertainty. In this paper, we propose a novel algorithm for constructing predicted classification confidences for DNNs that comes with provable correctness guarantees. Our approach uses Clopper-Pearson confidence intervals for the Binomial distribution in conjunction with the histogram binning approach to calibrated prediction. In addition, we demonstrate how our predicted confidences can be used to enable downstream guarantees in two settings: (i) fast "},"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":"2011.00716","kind":"arxiv","version":5},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2020-11-02T04:09:17Z","cross_cats_sorted":["stat.ML"],"title_canon_sha256":"4c3157a218bb9491fe947134ed00909e98a0d67081bd546053f4389085585e86","abstract_canon_sha256":"5d27735c521b0b8dc79b2ba328d5e44852656a6ca4b27469d6ffa26debb78546"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T02:24:11.783553Z","signature_b64":"DpPwH8NC7NsezDmaqTN1g8NtkvN8owZKfeIX25vWSpT+wWAMKbU37QsFFj6Oi1wIEYtq2PEtn0ptCCfMmmcsBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"c57feb88903ee3c6ff9714c46c9de3ecd801bdaeb2cc89ad3541689bddd37a2b","last_reissued_at":"2026-07-05T02:24:11.783130Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T02:24:11.783130Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"PAC Confidence Predictions for Deep Neural Network Classifiers","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Insup Lee, Osbert Bastani, Sangdon Park, Shuo Li","submitted_at":"2020-11-02T04:09:17Z","abstract_excerpt":"A key challenge for deploying deep neural networks (DNNs) in safety critical settings is the need to provide rigorous ways to quantify their uncertainty. In this paper, we propose a novel algorithm for constructing predicted classification confidences for DNNs that comes with provable correctness guarantees. Our approach uses Clopper-Pearson confidence intervals for the Binomial distribution in conjunction with the histogram binning approach to calibrated prediction. In addition, we demonstrate how our predicted confidences can be used to enable downstream guarantees in two settings: (i) fast "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2011.00716","kind":"arxiv","version":5},"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/2011.00716/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":"2011.00716","created_at":"2026-07-05T02:24:11.783186+00:00"},{"alias_kind":"arxiv_version","alias_value":"2011.00716v5","created_at":"2026-07-05T02:24:11.783186+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2011.00716","created_at":"2026-07-05T02:24:11.783186+00:00"},{"alias_kind":"pith_short_12","alias_value":"YV76XCEQH3R4","created_at":"2026-07-05T02:24:11.783186+00:00"},{"alias_kind":"pith_short_16","alias_value":"YV76XCEQH3R4N74X","created_at":"2026-07-05T02:24:11.783186+00:00"},{"alias_kind":"pith_short_8","alias_value":"YV76XCEQ","created_at":"2026-07-05T02:24:11.783186+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2506.03670","citing_title":"Position: There Is No Free Bayesian Uncertainty Quantification","ref_index":19,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/YV76XCEQH3R4N74XCTCGZHPD5T","json":"https://pith.science/pith/YV76XCEQH3R4N74XCTCGZHPD5T.json","graph_json":"https://pith.science/api/pith-number/YV76XCEQH3R4N74XCTCGZHPD5T/graph.json","events_json":"https://pith.science/api/pith-number/YV76XCEQH3R4N74XCTCGZHPD5T/events.json","paper":"https://pith.science/paper/YV76XCEQ"},"agent_actions":{"view_html":"https://pith.science/pith/YV76XCEQH3R4N74XCTCGZHPD5T","download_json":"https://pith.science/pith/YV76XCEQH3R4N74XCTCGZHPD5T.json","view_paper":"https://pith.science/paper/YV76XCEQ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2011.00716&json=true","fetch_graph":"https://pith.science/api/pith-number/YV76XCEQH3R4N74XCTCGZHPD5T/graph.json","fetch_events":"https://pith.science/api/pith-number/YV76XCEQH3R4N74XCTCGZHPD5T/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/YV76XCEQH3R4N74XCTCGZHPD5T/action/timestamp_anchor","attest_storage":"https://pith.science/pith/YV76XCEQH3R4N74XCTCGZHPD5T/action/storage_attestation","attest_author":"https://pith.science/pith/YV76XCEQH3R4N74XCTCGZHPD5T/action/author_attestation","sign_citation":"https://pith.science/pith/YV76XCEQH3R4N74XCTCGZHPD5T/action/citation_signature","submit_replication":"https://pith.science/pith/YV76XCEQH3R4N74XCTCGZHPD5T/action/replication_record"}},"created_at":"2026-07-05T02:24:11.783186+00:00","updated_at":"2026-07-05T02:24:11.783186+00:00"}