{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:MDNCOIHICUSMIURFM2OVQW4C64","short_pith_number":"pith:MDNCOIHI","schema_version":"1.0","canonical_sha256":"60da2720e81524c45225669d585b82f71e4a053d50bc8246828155695f1f0e5e","source":{"kind":"arxiv","id":"2110.11216","version":6},"attestation_state":"computed","paper":{"title":"User-friendly introduction to PAC-Bayes bounds","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG","math.ST","stat.TH"],"primary_cat":"stat.ML","authors_text":"Pierre Alquier","submitted_at":"2021-10-21T15:50:05Z","abstract_excerpt":"Aggregated predictors are obtained by making a set of basic predictors vote according to some weights, that is, to some probability distribution.\n  Randomized predictors are obtained by sampling in a set of basic predictors, according to some prescribed probability distribution.\n  Thus, aggregated and randomized predictors have in common that they are not defined by a minimization problem, but by a probability distribution on the set of predictors. In statistical learning theory, there is a set of tools designed to understand the generalization ability of such procedures: PAC-Bayesian or PAC-B"},"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":"2110.11216","kind":"arxiv","version":6},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"stat.ML","submitted_at":"2021-10-21T15:50:05Z","cross_cats_sorted":["cs.LG","math.ST","stat.TH"],"title_canon_sha256":"6f7680265018a467d8debd5f6ac687d0aca9a4ec6c69fb622ee9cdddfa3d2835","abstract_canon_sha256":"b35c8fdd8e151ff4be171571746e6096e23b6b213de06d9dd42d384191202449"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:21:14.961378Z","signature_b64":"QjYE0Yc0RDYpOxm9kNBr9hs5Ydx/cFHiFakYoCppy//aunhvYHelqMx9qlMHCA8SObin0GXrPY+JsWRBY9YUBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"60da2720e81524c45225669d585b82f71e4a053d50bc8246828155695f1f0e5e","last_reissued_at":"2026-07-05T10:21:14.960801Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:21:14.960801Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"User-friendly introduction to PAC-Bayes bounds","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG","math.ST","stat.TH"],"primary_cat":"stat.ML","authors_text":"Pierre Alquier","submitted_at":"2021-10-21T15:50:05Z","abstract_excerpt":"Aggregated predictors are obtained by making a set of basic predictors vote according to some weights, that is, to some probability distribution.\n  Randomized predictors are obtained by sampling in a set of basic predictors, according to some prescribed probability distribution.\n  Thus, aggregated and randomized predictors have in common that they are not defined by a minimization problem, but by a probability distribution on the set of predictors. In statistical learning theory, there is a set of tools designed to understand the generalization ability of such procedures: PAC-Bayesian or PAC-B"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2110.11216","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":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2110.11216/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":"2110.11216","created_at":"2026-07-05T10:21:14.960862+00:00"},{"alias_kind":"arxiv_version","alias_value":"2110.11216v6","created_at":"2026-07-05T10:21:14.960862+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2110.11216","created_at":"2026-07-05T10:21:14.960862+00:00"},{"alias_kind":"pith_short_12","alias_value":"MDNCOIHICUSM","created_at":"2026-07-05T10:21:14.960862+00:00"},{"alias_kind":"pith_short_16","alias_value":"MDNCOIHICUSMIURF","created_at":"2026-07-05T10:21:14.960862+00:00"},{"alias_kind":"pith_short_8","alias_value":"MDNCOIHI","created_at":"2026-07-05T10:21:14.960862+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":12,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.23668","citing_title":"On the Limits of Prompt-Conditioned Language Models as General-Purpose Learners","ref_index":1,"is_internal_anchor":false},{"citing_arxiv_id":"2607.02247","citing_title":"Aggregation with Exponential Weights is Optimal in Expectation","ref_index":1,"is_internal_anchor":false},{"citing_arxiv_id":"2606.28281","citing_title":"PAC-Bayesian Certificates for Quadratic Closed-Loop Control","ref_index":1,"is_internal_anchor":false},{"citing_arxiv_id":"2606.29331","citing_title":"Sample Complexity of Scientific Discovery: PAC Learnability of Compositional Function Trees","ref_index":44,"is_internal_anchor":false},{"citing_arxiv_id":"2605.25612","citing_title":"Towards the Connection between Activation Sparsity and Flat Minima","ref_index":51,"is_internal_anchor":false},{"citing_arxiv_id":"2202.05568","citing_title":"Change of measure through the Legendre transform","ref_index":1,"is_internal_anchor":false},{"citing_arxiv_id":"2310.11203","citing_title":"Federated Learning with Nonvacuous Generalisation Bounds","ref_index":3,"is_internal_anchor":false},{"citing_arxiv_id":"2602.07999","citing_title":"Tighter Information-Theoretic Generalization Bounds via a Novel Class of Change of Measure Inequalities","ref_index":45,"is_internal_anchor":false},{"citing_arxiv_id":"2605.11170","citing_title":"Unlearning with Asymmetric Sources: Improved Unlearning-Utility Trade-off with Public Data","ref_index":100,"is_internal_anchor":false},{"citing_arxiv_id":"2605.09471","citing_title":"The Statistical Cost of Adaptation in Multi-Source Transfer Learning","ref_index":154,"is_internal_anchor":false},{"citing_arxiv_id":"2604.10588","citing_title":"Distributionally Robust PAC-Bayesian Control","ref_index":17,"is_internal_anchor":false},{"citing_arxiv_id":"2604.17219","citing_title":"PAC-Bayes Bounds for Gibbs Posteriors via Singular Learning Theory","ref_index":15,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/MDNCOIHICUSMIURFM2OVQW4C64","json":"https://pith.science/pith/MDNCOIHICUSMIURFM2OVQW4C64.json","graph_json":"https://pith.science/api/pith-number/MDNCOIHICUSMIURFM2OVQW4C64/graph.json","events_json":"https://pith.science/api/pith-number/MDNCOIHICUSMIURFM2OVQW4C64/events.json","paper":"https://pith.science/paper/MDNCOIHI"},"agent_actions":{"view_html":"https://pith.science/pith/MDNCOIHICUSMIURFM2OVQW4C64","download_json":"https://pith.science/pith/MDNCOIHICUSMIURFM2OVQW4C64.json","view_paper":"https://pith.science/paper/MDNCOIHI","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2110.11216&json=true","fetch_graph":"https://pith.science/api/pith-number/MDNCOIHICUSMIURFM2OVQW4C64/graph.json","fetch_events":"https://pith.science/api/pith-number/MDNCOIHICUSMIURFM2OVQW4C64/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/MDNCOIHICUSMIURFM2OVQW4C64/action/timestamp_anchor","attest_storage":"https://pith.science/pith/MDNCOIHICUSMIURFM2OVQW4C64/action/storage_attestation","attest_author":"https://pith.science/pith/MDNCOIHICUSMIURFM2OVQW4C64/action/author_attestation","sign_citation":"https://pith.science/pith/MDNCOIHICUSMIURFM2OVQW4C64/action/citation_signature","submit_replication":"https://pith.science/pith/MDNCOIHICUSMIURFM2OVQW4C64/action/replication_record"}},"created_at":"2026-07-05T10:21:14.960862+00:00","updated_at":"2026-07-05T10:21:14.960862+00:00"}