{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2010:D2ZRYHWCKTVZYVWQBL44P276VZ","short_pith_number":"pith:D2ZRYHWC","schema_version":"1.0","canonical_sha256":"1eb31c1ec254eb9c56d00af9c7ebfeae4eb93e4f22c5fe302f4658b0007df2e5","source":{"kind":"arxiv","id":"1004.5229","version":3},"attestation_state":"computed","paper":{"title":"Optimism in Reinforcement Learning and Kullback-Leibler Divergence","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["math.ST","stat.ML","stat.TH"],"primary_cat":"cs.LG","authors_text":"Aur\\'elien Garivier (LTCI), Olivier Capp\\'e (LTCI), Sarah Filippi (LTCI)","submitted_at":"2010-04-29T09:31:55Z","abstract_excerpt":"We consider model-based reinforcement learning in finite Markov De- cision Processes (MDPs), focussing on so-called optimistic strategies. In MDPs, optimism can be implemented by carrying out extended value it- erations under a constraint of consistency with the estimated model tran- sition probabilities. The UCRL2 algorithm by Auer, Jaksch and Ortner (2009), which follows this strategy, has recently been shown to guarantee near-optimal regret bounds. In this paper, we strongly argue in favor of using the Kullback-Leibler (KL) divergence for this purpose. By studying the linear maximization pr"},"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":"1004.5229","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2010-04-29T09:31:55Z","cross_cats_sorted":["math.ST","stat.ML","stat.TH"],"title_canon_sha256":"ea0a9ccfe89bebb1299f58dd84d78444355348965483bd1abffae536239f3e91","abstract_canon_sha256":"0459926e27a62e60dd7ce51e657a8f589cabff0e3bd00672a73cfc19c7e5d8ff"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-18T04:12:43.062885Z","signature_b64":"KdanyIaHT6m3wKVtmY+YWO1kfCwlsu6sufVAYzo2VdEQaY1xf9cSZpUordf398bP0HawNqwOpUJFgqKF163HBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"1eb31c1ec254eb9c56d00af9c7ebfeae4eb93e4f22c5fe302f4658b0007df2e5","last_reissued_at":"2026-05-18T04:12:43.062146Z","signature_status":"signed_v1","first_computed_at":"2026-05-18T04:12:43.062146Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Optimism in Reinforcement Learning and Kullback-Leibler Divergence","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["math.ST","stat.ML","stat.TH"],"primary_cat":"cs.LG","authors_text":"Aur\\'elien Garivier (LTCI), Olivier Capp\\'e (LTCI), Sarah Filippi (LTCI)","submitted_at":"2010-04-29T09:31:55Z","abstract_excerpt":"We consider model-based reinforcement learning in finite Markov De- cision Processes (MDPs), focussing on so-called optimistic strategies. In MDPs, optimism can be implemented by carrying out extended value it- erations under a constraint of consistency with the estimated model tran- sition probabilities. The UCRL2 algorithm by Auer, Jaksch and Ortner (2009), which follows this strategy, has recently been shown to guarantee near-optimal regret bounds. In this paper, we strongly argue in favor of using the Kullback-Leibler (KL) divergence for this purpose. By studying the linear maximization pr"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1004.5229","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":""},"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":"1004.5229","created_at":"2026-05-18T04:12:43.062237+00:00"},{"alias_kind":"arxiv_version","alias_value":"1004.5229v3","created_at":"2026-05-18T04:12:43.062237+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1004.5229","created_at":"2026-05-18T04:12:43.062237+00:00"},{"alias_kind":"pith_short_12","alias_value":"D2ZRYHWCKTVZ","created_at":"2026-05-18T12:26:06.534383+00:00"},{"alias_kind":"pith_short_16","alias_value":"D2ZRYHWCKTVZYVWQ","created_at":"2026-05-18T12:26:06.534383+00:00"},{"alias_kind":"pith_short_8","alias_value":"D2ZRYHWC","created_at":"2026-05-18T12:26:06.534383+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/D2ZRYHWCKTVZYVWQBL44P276VZ","json":"https://pith.science/pith/D2ZRYHWCKTVZYVWQBL44P276VZ.json","graph_json":"https://pith.science/api/pith-number/D2ZRYHWCKTVZYVWQBL44P276VZ/graph.json","events_json":"https://pith.science/api/pith-number/D2ZRYHWCKTVZYVWQBL44P276VZ/events.json","paper":"https://pith.science/paper/D2ZRYHWC"},"agent_actions":{"view_html":"https://pith.science/pith/D2ZRYHWCKTVZYVWQBL44P276VZ","download_json":"https://pith.science/pith/D2ZRYHWCKTVZYVWQBL44P276VZ.json","view_paper":"https://pith.science/paper/D2ZRYHWC","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1004.5229&json=true","fetch_graph":"https://pith.science/api/pith-number/D2ZRYHWCKTVZYVWQBL44P276VZ/graph.json","fetch_events":"https://pith.science/api/pith-number/D2ZRYHWCKTVZYVWQBL44P276VZ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/D2ZRYHWCKTVZYVWQBL44P276VZ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/D2ZRYHWCKTVZYVWQBL44P276VZ/action/storage_attestation","attest_author":"https://pith.science/pith/D2ZRYHWCKTVZYVWQBL44P276VZ/action/author_attestation","sign_citation":"https://pith.science/pith/D2ZRYHWCKTVZYVWQBL44P276VZ/action/citation_signature","submit_replication":"https://pith.science/pith/D2ZRYHWCKTVZYVWQBL44P276VZ/action/replication_record"}},"created_at":"2026-05-18T04:12:43.062237+00:00","updated_at":"2026-05-18T04:12:43.062237+00:00"}