{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:MGEBWEYWR4S3PTYNVE5R5LD5NZ","short_pith_number":"pith:MGEBWEYW","schema_version":"1.0","canonical_sha256":"61881b13168f25b7cf0da93b1eac7d6e5cb656b9bca7f6a64ca23e48ed22ba17","source":{"kind":"arxiv","id":"2405.17108","version":1},"attestation_state":"computed","paper":{"title":"Finding good policies in average-reward Markov Decision Processes without prior knowledge","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Adrienne Tuynman, Emilie Kaufmann, R\\'emy Degenne","submitted_at":"2024-05-27T12:24:14Z","abstract_excerpt":"We revisit the identification of an $\\varepsilon$-optimal policy in average-reward Markov Decision Processes (MDP). In such MDPs, two measures of complexity have appeared in the literature: the diameter, $D$, and the optimal bias span, $H$, which satisfy $H\\leq D$. Prior work have studied the complexity of $\\varepsilon$-optimal policy identification only when a generative model is available. In this case, it is known that there exists an MDP with $D \\simeq H$ for which the sample complexity to output an $\\varepsilon$-optimal policy is $\\Omega(SAD/\\varepsilon^2)$ where $S$ and $A$ are the sizes"},"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":"2405.17108","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-05-27T12:24:14Z","cross_cats_sorted":[],"title_canon_sha256":"740b87900539dd6a45ccda315eaa00bd69d9f99a9e46d8611610e64d5d33a318","abstract_canon_sha256":"a3f203ff76852a21ef1028fb9e0c8d3e080aa8886021f5243ff279edc3c85b3f"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:23:40.077764Z","signature_b64":"WvSzKpvpZBe1GqzKfRmUb/ZIzA7Wa9EkHj/OrwJnnaI8sCd042ykek95VBke981dj8aPhi+cfh1MU6nZkUnJAg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"61881b13168f25b7cf0da93b1eac7d6e5cb656b9bca7f6a64ca23e48ed22ba17","last_reissued_at":"2026-07-05T08:23:40.077216Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:23:40.077216Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Finding good policies in average-reward Markov Decision Processes without prior knowledge","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Adrienne Tuynman, Emilie Kaufmann, R\\'emy Degenne","submitted_at":"2024-05-27T12:24:14Z","abstract_excerpt":"We revisit the identification of an $\\varepsilon$-optimal policy in average-reward Markov Decision Processes (MDP). In such MDPs, two measures of complexity have appeared in the literature: the diameter, $D$, and the optimal bias span, $H$, which satisfy $H\\leq D$. Prior work have studied the complexity of $\\varepsilon$-optimal policy identification only when a generative model is available. In this case, it is known that there exists an MDP with $D \\simeq H$ for which the sample complexity to output an $\\varepsilon$-optimal policy is $\\Omega(SAD/\\varepsilon^2)$ where $S$ and $A$ are the sizes"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2405.17108","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/2405.17108/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":"2405.17108","created_at":"2026-07-05T08:23:40.077286+00:00"},{"alias_kind":"arxiv_version","alias_value":"2405.17108v1","created_at":"2026-07-05T08:23:40.077286+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2405.17108","created_at":"2026-07-05T08:23:40.077286+00:00"},{"alias_kind":"pith_short_12","alias_value":"MGEBWEYWR4S3","created_at":"2026-07-05T08:23:40.077286+00:00"},{"alias_kind":"pith_short_16","alias_value":"MGEBWEYWR4S3PTYN","created_at":"2026-07-05T08:23:40.077286+00:00"},{"alias_kind":"pith_short_8","alias_value":"MGEBWEYW","created_at":"2026-07-05T08:23:40.077286+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2505.12462","citing_title":"Model-Free Robust Average-Reward Reinforcement Learning with Sample Complexity Analysis","ref_index":75,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/MGEBWEYWR4S3PTYNVE5R5LD5NZ","json":"https://pith.science/pith/MGEBWEYWR4S3PTYNVE5R5LD5NZ.json","graph_json":"https://pith.science/api/pith-number/MGEBWEYWR4S3PTYNVE5R5LD5NZ/graph.json","events_json":"https://pith.science/api/pith-number/MGEBWEYWR4S3PTYNVE5R5LD5NZ/events.json","paper":"https://pith.science/paper/MGEBWEYW"},"agent_actions":{"view_html":"https://pith.science/pith/MGEBWEYWR4S3PTYNVE5R5LD5NZ","download_json":"https://pith.science/pith/MGEBWEYWR4S3PTYNVE5R5LD5NZ.json","view_paper":"https://pith.science/paper/MGEBWEYW","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2405.17108&json=true","fetch_graph":"https://pith.science/api/pith-number/MGEBWEYWR4S3PTYNVE5R5LD5NZ/graph.json","fetch_events":"https://pith.science/api/pith-number/MGEBWEYWR4S3PTYNVE5R5LD5NZ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/MGEBWEYWR4S3PTYNVE5R5LD5NZ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/MGEBWEYWR4S3PTYNVE5R5LD5NZ/action/storage_attestation","attest_author":"https://pith.science/pith/MGEBWEYWR4S3PTYNVE5R5LD5NZ/action/author_attestation","sign_citation":"https://pith.science/pith/MGEBWEYWR4S3PTYNVE5R5LD5NZ/action/citation_signature","submit_replication":"https://pith.science/pith/MGEBWEYWR4S3PTYNVE5R5LD5NZ/action/replication_record"}},"created_at":"2026-07-05T08:23:40.077286+00:00","updated_at":"2026-07-05T08:23:40.077286+00:00"}