{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:F7MDY26QDEGS5C4J54YVYLKUF2","short_pith_number":"pith:F7MDY26Q","schema_version":"1.0","canonical_sha256":"2fd83c6bd0190d2e8b89ef315c2d542e92baf0ff937efda526893945437958d9","source":{"kind":"arxiv","id":"2407.13977","version":3},"attestation_state":"computed","paper":{"title":"A Unified Confidence Sequence for Generalized Linear Models, with Applications to Bandits","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"stat.ML","authors_text":"Junghyun Lee, Kwang-Sung Jun, Se-Young Yun","submitted_at":"2024-07-19T02:06:08Z","abstract_excerpt":"We present a unified likelihood ratio-based confidence sequence (CS) for any (self-concordant) generalized linear model (GLM) that is guaranteed to be convex and numerically tight. We show that this is on par or improves upon known CSs for various GLMs, including Gaussian, Bernoulli, and Poisson. In particular, for the first time, our CS for Bernoulli has a $\\mathrm{poly}(S)$-free radius where $S$ is the norm of the unknown parameter. Our first technical novelty is its derivation, which utilizes a time-uniform PAC-Bayesian bound with a uniform prior/posterior, despite the latter being a rather"},"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":"2407.13977","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"stat.ML","submitted_at":"2024-07-19T02:06:08Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"a5eb478c5ac6483b8b76e7c5730df7d7c1695910a8da97a5ba0c488358e6b4cf","abstract_canon_sha256":"2d3996ae65c8bec9e443848fb026b579467b45c68b2376ab553f91a514409bc1"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:01:14.928612Z","signature_b64":"hxseOHR637oON49LW0uhKe+CAyfILSpbjfrP9OeuxkO98f8rpCC30Xiddo/hKZfCcTBs3rLvQFDdMEZjFtTnBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"2fd83c6bd0190d2e8b89ef315c2d542e92baf0ff937efda526893945437958d9","last_reissued_at":"2026-07-05T10:01:14.928055Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:01:14.928055Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"A Unified Confidence Sequence for Generalized Linear Models, with Applications to Bandits","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"stat.ML","authors_text":"Junghyun Lee, Kwang-Sung Jun, Se-Young Yun","submitted_at":"2024-07-19T02:06:08Z","abstract_excerpt":"We present a unified likelihood ratio-based confidence sequence (CS) for any (self-concordant) generalized linear model (GLM) that is guaranteed to be convex and numerically tight. We show that this is on par or improves upon known CSs for various GLMs, including Gaussian, Bernoulli, and Poisson. In particular, for the first time, our CS for Bernoulli has a $\\mathrm{poly}(S)$-free radius where $S$ is the norm of the unknown parameter. Our first technical novelty is its derivation, which utilizes a time-uniform PAC-Bayesian bound with a uniform prior/posterior, despite the latter being a rather"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2407.13977","kind":"arxiv","version":3},"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/2407.13977/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":"2407.13977","created_at":"2026-07-05T10:01:14.928117+00:00"},{"alias_kind":"arxiv_version","alias_value":"2407.13977v3","created_at":"2026-07-05T10:01:14.928117+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2407.13977","created_at":"2026-07-05T10:01:14.928117+00:00"},{"alias_kind":"pith_short_12","alias_value":"F7MDY26QDEGS","created_at":"2026-07-05T10:01:14.928117+00:00"},{"alias_kind":"pith_short_16","alias_value":"F7MDY26QDEGS5C4J","created_at":"2026-07-05T10:01:14.928117+00:00"},{"alias_kind":"pith_short_8","alias_value":"F7MDY26Q","created_at":"2026-07-05T10:01:14.928117+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":3,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.09668","citing_title":"Algorithm for Contextual Queueing Bandits with Rate-Optimal Queue Length Regret","ref_index":6,"is_internal_anchor":false},{"citing_arxiv_id":"2605.05745","citing_title":"Best Arm Identification in Generalized Linear Bandits via Hybrid Feedback","ref_index":4,"is_internal_anchor":false},{"citing_arxiv_id":"2604.19008","citing_title":"Optimal Online and Offline Algorithms for Contextual MNL with Applications to Assortment and Pricing","ref_index":25,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/F7MDY26QDEGS5C4J54YVYLKUF2","json":"https://pith.science/pith/F7MDY26QDEGS5C4J54YVYLKUF2.json","graph_json":"https://pith.science/api/pith-number/F7MDY26QDEGS5C4J54YVYLKUF2/graph.json","events_json":"https://pith.science/api/pith-number/F7MDY26QDEGS5C4J54YVYLKUF2/events.json","paper":"https://pith.science/paper/F7MDY26Q"},"agent_actions":{"view_html":"https://pith.science/pith/F7MDY26QDEGS5C4J54YVYLKUF2","download_json":"https://pith.science/pith/F7MDY26QDEGS5C4J54YVYLKUF2.json","view_paper":"https://pith.science/paper/F7MDY26Q","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2407.13977&json=true","fetch_graph":"https://pith.science/api/pith-number/F7MDY26QDEGS5C4J54YVYLKUF2/graph.json","fetch_events":"https://pith.science/api/pith-number/F7MDY26QDEGS5C4J54YVYLKUF2/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/F7MDY26QDEGS5C4J54YVYLKUF2/action/timestamp_anchor","attest_storage":"https://pith.science/pith/F7MDY26QDEGS5C4J54YVYLKUF2/action/storage_attestation","attest_author":"https://pith.science/pith/F7MDY26QDEGS5C4J54YVYLKUF2/action/author_attestation","sign_citation":"https://pith.science/pith/F7MDY26QDEGS5C4J54YVYLKUF2/action/citation_signature","submit_replication":"https://pith.science/pith/F7MDY26QDEGS5C4J54YVYLKUF2/action/replication_record"}},"created_at":"2026-07-05T10:01:14.928117+00:00","updated_at":"2026-07-05T10:01:14.928117+00:00"}