{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2019:WL5DQIES5C5CBJWWCRRMH2PYRO","short_pith_number":"pith:WL5DQIES","schema_version":"1.0","canonical_sha256":"b2fa382092e8ba20a6d61462c3e9f88b97e7064226c34f9ee3d4d876fef64081","source":{"kind":"arxiv","id":"1902.00746","version":3},"attestation_state":"computed","paper":{"title":"On the bias, risk and consistency of sample means in multi-armed bandits","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG","stat.ML","stat.TH"],"primary_cat":"math.ST","authors_text":"Aaditya Ramdas, Alessandro Rinaldo, Jaehyeok Shin","submitted_at":"2019-02-02T16:23:08Z","abstract_excerpt":"The sample mean is among the most well studied estimators in statistics, having many desirable properties such as unbiasedness and consistency. However, when analyzing data collected using a multi-armed bandit (MAB) experiment, the sample mean is biased and much remains to be understood about its properties. For example, when is it consistent, how large is its bias, and can we bound its mean squared error? This paper delivers a thorough and systematic treatment of the bias, risk and consistency of MAB sample means. Specifically, we identify four distinct sources of selection bias (sampling, st"},"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":"1902.00746","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"math.ST","submitted_at":"2019-02-02T16:23:08Z","cross_cats_sorted":["cs.LG","stat.ML","stat.TH"],"title_canon_sha256":"f85ecb367bd8e332d3bbf2da3dc5e38701f1d85d6c494d96119707a50fd9c4af","abstract_canon_sha256":"19172097c4c26f1816c17727609957811e48eca42ec2e8b0b56654eaf48b55a5"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T02:36:18.466497Z","signature_b64":"/QzKjELCFzWTZ6f/Sb6uc/X1PSnP2aVKaaaoZwloqmYqEXLHjcxHoDmihmrtp5HoFqkj2jseU8E97VYX3ZgDDg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"b2fa382092e8ba20a6d61462c3e9f88b97e7064226c34f9ee3d4d876fef64081","last_reissued_at":"2026-07-05T02:36:18.466064Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T02:36:18.466064Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"On the bias, risk and consistency of sample means in multi-armed bandits","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG","stat.ML","stat.TH"],"primary_cat":"math.ST","authors_text":"Aaditya Ramdas, Alessandro Rinaldo, Jaehyeok Shin","submitted_at":"2019-02-02T16:23:08Z","abstract_excerpt":"The sample mean is among the most well studied estimators in statistics, having many desirable properties such as unbiasedness and consistency. However, when analyzing data collected using a multi-armed bandit (MAB) experiment, the sample mean is biased and much remains to be understood about its properties. For example, when is it consistent, how large is its bias, and can we bound its mean squared error? This paper delivers a thorough and systematic treatment of the bias, risk and consistency of MAB sample means. Specifically, we identify four distinct sources of selection bias (sampling, st"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1902.00746","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/1902.00746/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":"1902.00746","created_at":"2026-07-05T02:36:18.466115+00:00"},{"alias_kind":"arxiv_version","alias_value":"1902.00746v3","created_at":"2026-07-05T02:36:18.466115+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1902.00746","created_at":"2026-07-05T02:36:18.466115+00:00"},{"alias_kind":"pith_short_12","alias_value":"WL5DQIES5C5C","created_at":"2026-07-05T02:36:18.466115+00:00"},{"alias_kind":"pith_short_16","alias_value":"WL5DQIES5C5CBJWW","created_at":"2026-07-05T02:36:18.466115+00:00"},{"alias_kind":"pith_short_8","alias_value":"WL5DQIES","created_at":"2026-07-05T02:36:18.466115+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.16615","citing_title":"Learning What Evaluators Value: A Reliable Approach to Modeling Evaluator Preferences","ref_index":43,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/WL5DQIES5C5CBJWWCRRMH2PYRO","json":"https://pith.science/pith/WL5DQIES5C5CBJWWCRRMH2PYRO.json","graph_json":"https://pith.science/api/pith-number/WL5DQIES5C5CBJWWCRRMH2PYRO/graph.json","events_json":"https://pith.science/api/pith-number/WL5DQIES5C5CBJWWCRRMH2PYRO/events.json","paper":"https://pith.science/paper/WL5DQIES"},"agent_actions":{"view_html":"https://pith.science/pith/WL5DQIES5C5CBJWWCRRMH2PYRO","download_json":"https://pith.science/pith/WL5DQIES5C5CBJWWCRRMH2PYRO.json","view_paper":"https://pith.science/paper/WL5DQIES","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1902.00746&json=true","fetch_graph":"https://pith.science/api/pith-number/WL5DQIES5C5CBJWWCRRMH2PYRO/graph.json","fetch_events":"https://pith.science/api/pith-number/WL5DQIES5C5CBJWWCRRMH2PYRO/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/WL5DQIES5C5CBJWWCRRMH2PYRO/action/timestamp_anchor","attest_storage":"https://pith.science/pith/WL5DQIES5C5CBJWWCRRMH2PYRO/action/storage_attestation","attest_author":"https://pith.science/pith/WL5DQIES5C5CBJWWCRRMH2PYRO/action/author_attestation","sign_citation":"https://pith.science/pith/WL5DQIES5C5CBJWWCRRMH2PYRO/action/citation_signature","submit_replication":"https://pith.science/pith/WL5DQIES5C5CBJWWCRRMH2PYRO/action/replication_record"}},"created_at":"2026-07-05T02:36:18.466115+00:00","updated_at":"2026-07-05T02:36:18.466115+00:00"}