{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:UDGRBDVGA7BY6D6ENENQHV3H7Q","short_pith_number":"pith:UDGRBDVG","schema_version":"1.0","canonical_sha256":"a0cd108ea607c38f0fc4691b03d767fc2e8b5009b90b5248ebb14634afe59f1f","source":{"kind":"arxiv","id":"2202.01454","version":1},"attestation_state":"computed","paper":{"title":"Deep Hierarchy in Bandits","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Branislav Kveton, Joey Hong, Manzil Zaheer, Mohammad Ghavamzadeh, Sumeet Katariya","submitted_at":"2022-02-03T08:15:53Z","abstract_excerpt":"Mean rewards of actions are often correlated. The form of these correlations may be complex and unknown a priori, such as the preferences of a user for recommended products and their categories. To maximize statistical efficiency, it is important to leverage these correlations when learning. We formulate a bandit variant of this problem where the correlations of mean action rewards are represented by a hierarchical Bayesian model with latent variables. Since the hierarchy can have multiple layers, we call it deep. We propose a hierarchical Thompson sampling algorithm (HierTS) for this problem,"},"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":"2202.01454","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2022-02-03T08:15:53Z","cross_cats_sorted":["stat.ML"],"title_canon_sha256":"78d618e27292078da2663dc69041689f74a683e518cfc21c5edb0a1cde61ab06","abstract_canon_sha256":"2fa2eb2cb69f9e3a4489c65ac04321c5794033ccc8a7738b96670b1a73427727"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T03:53:54.466128Z","signature_b64":"S15wsRXBucw8jNRL/FMGQcxEV1MBmHY8RGqJZu52wCiZ8e3gHrbTto0VQ+0Ch+u/JofCtVm2MVQPDIzCXd5jBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"a0cd108ea607c38f0fc4691b03d767fc2e8b5009b90b5248ebb14634afe59f1f","last_reissued_at":"2026-07-05T03:53:54.465688Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T03:53:54.465688Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Deep Hierarchy in Bandits","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Branislav Kveton, Joey Hong, Manzil Zaheer, Mohammad Ghavamzadeh, Sumeet Katariya","submitted_at":"2022-02-03T08:15:53Z","abstract_excerpt":"Mean rewards of actions are often correlated. The form of these correlations may be complex and unknown a priori, such as the preferences of a user for recommended products and their categories. To maximize statistical efficiency, it is important to leverage these correlations when learning. We formulate a bandit variant of this problem where the correlations of mean action rewards are represented by a hierarchical Bayesian model with latent variables. Since the hierarchy can have multiple layers, we call it deep. We propose a hierarchical Thompson sampling algorithm (HierTS) for this problem,"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2202.01454","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":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2202.01454/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":"2202.01454","created_at":"2026-07-05T03:53:54.465753+00:00"},{"alias_kind":"arxiv_version","alias_value":"2202.01454v1","created_at":"2026-07-05T03:53:54.465753+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2202.01454","created_at":"2026-07-05T03:53:54.465753+00:00"},{"alias_kind":"pith_short_12","alias_value":"UDGRBDVGA7BY","created_at":"2026-07-05T03:53:54.465753+00:00"},{"alias_kind":"pith_short_16","alias_value":"UDGRBDVGA7BY6D6E","created_at":"2026-07-05T03:53:54.465753+00:00"},{"alias_kind":"pith_short_8","alias_value":"UDGRBDVG","created_at":"2026-07-05T03:53:54.465753+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/UDGRBDVGA7BY6D6ENENQHV3H7Q","json":"https://pith.science/pith/UDGRBDVGA7BY6D6ENENQHV3H7Q.json","graph_json":"https://pith.science/api/pith-number/UDGRBDVGA7BY6D6ENENQHV3H7Q/graph.json","events_json":"https://pith.science/api/pith-number/UDGRBDVGA7BY6D6ENENQHV3H7Q/events.json","paper":"https://pith.science/paper/UDGRBDVG"},"agent_actions":{"view_html":"https://pith.science/pith/UDGRBDVGA7BY6D6ENENQHV3H7Q","download_json":"https://pith.science/pith/UDGRBDVGA7BY6D6ENENQHV3H7Q.json","view_paper":"https://pith.science/paper/UDGRBDVG","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2202.01454&json=true","fetch_graph":"https://pith.science/api/pith-number/UDGRBDVGA7BY6D6ENENQHV3H7Q/graph.json","fetch_events":"https://pith.science/api/pith-number/UDGRBDVGA7BY6D6ENENQHV3H7Q/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/UDGRBDVGA7BY6D6ENENQHV3H7Q/action/timestamp_anchor","attest_storage":"https://pith.science/pith/UDGRBDVGA7BY6D6ENENQHV3H7Q/action/storage_attestation","attest_author":"https://pith.science/pith/UDGRBDVGA7BY6D6ENENQHV3H7Q/action/author_attestation","sign_citation":"https://pith.science/pith/UDGRBDVGA7BY6D6ENENQHV3H7Q/action/citation_signature","submit_replication":"https://pith.science/pith/UDGRBDVGA7BY6D6ENENQHV3H7Q/action/replication_record"}},"created_at":"2026-07-05T03:53:54.465753+00:00","updated_at":"2026-07-05T03:53:54.465753+00:00"}