{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:DIQJ7VWGI5ZFURS6IIJCGODEFH","short_pith_number":"pith:DIQJ7VWG","schema_version":"1.0","canonical_sha256":"1a209fd6c647725a465e421223386429fc44c9f59c3c7bd036d39a8ced730440","source":{"kind":"arxiv","id":"2502.04477","version":2},"attestation_state":"computed","paper":{"title":"Near-Optimal Sample Complexity for MDPs via Anchoring","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.DS"],"primary_cat":"math.OC","authors_text":"Jongmin Lee, Mario Bravo, Roberto Cominetti","submitted_at":"2025-02-06T20:01:59Z","abstract_excerpt":"We study a new model-free algorithm to compute $\\varepsilon$-optimal policies for average reward Markov decision processes, in the weakly communicating case. Given a generative model, our procedure combines a recursive sampling technique with Halpern's anchored iteration, and computes an $\\varepsilon$-optimal policy with sample and time complexity $\\widetilde{O}(|\\mathcal{S}||\\mathcal{A}|\\|h^*\\|_{\\text{sp}}^{2}/\\varepsilon^{2})$ both in high probability and in expectation. To our knowledge, this is the best complexity among model-free algorithms, matching the known lower bound up to a factor $"},"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":"2502.04477","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"math.OC","submitted_at":"2025-02-06T20:01:59Z","cross_cats_sorted":["cs.DS"],"title_canon_sha256":"7896ced173e6a3d004eb9fc23958b750c92960a50701eb41c356767fea5d8bda","abstract_canon_sha256":"ee1abdbca0bb20c9454c2de4e3ed360f94287af4007f85a3c4c843becd4bcb15"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:20:50.377250Z","signature_b64":"ONlUJPhkCDrCmBHuxibYcMt2IprhFPT/ZpN1J3Fe5Ogw/+/B8YBtqpdPjf84PwDK/6pYtYRcrw8RT11XcxrcBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"1a209fd6c647725a465e421223386429fc44c9f59c3c7bd036d39a8ced730440","last_reissued_at":"2026-07-05T11:20:50.376780Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:20:50.376780Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Near-Optimal Sample Complexity for MDPs via Anchoring","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.DS"],"primary_cat":"math.OC","authors_text":"Jongmin Lee, Mario Bravo, Roberto Cominetti","submitted_at":"2025-02-06T20:01:59Z","abstract_excerpt":"We study a new model-free algorithm to compute $\\varepsilon$-optimal policies for average reward Markov decision processes, in the weakly communicating case. Given a generative model, our procedure combines a recursive sampling technique with Halpern's anchored iteration, and computes an $\\varepsilon$-optimal policy with sample and time complexity $\\widetilde{O}(|\\mathcal{S}||\\mathcal{A}|\\|h^*\\|_{\\text{sp}}^{2}/\\varepsilon^{2})$ both in high probability and in expectation. To our knowledge, this is the best complexity among model-free algorithms, matching the known lower bound up to a factor $"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2502.04477","kind":"arxiv","version":2},"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/2502.04477/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":"2502.04477","created_at":"2026-07-05T11:20:50.376838+00:00"},{"alias_kind":"arxiv_version","alias_value":"2502.04477v2","created_at":"2026-07-05T11:20:50.376838+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2502.04477","created_at":"2026-07-05T11:20:50.376838+00:00"},{"alias_kind":"pith_short_12","alias_value":"DIQJ7VWGI5ZF","created_at":"2026-07-05T11:20:50.376838+00:00"},{"alias_kind":"pith_short_16","alias_value":"DIQJ7VWGI5ZFURS6","created_at":"2026-07-05T11:20:50.376838+00:00"},{"alias_kind":"pith_short_8","alias_value":"DIQJ7VWG","created_at":"2026-07-05T11:20:50.376838+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/DIQJ7VWGI5ZFURS6IIJCGODEFH","json":"https://pith.science/pith/DIQJ7VWGI5ZFURS6IIJCGODEFH.json","graph_json":"https://pith.science/api/pith-number/DIQJ7VWGI5ZFURS6IIJCGODEFH/graph.json","events_json":"https://pith.science/api/pith-number/DIQJ7VWGI5ZFURS6IIJCGODEFH/events.json","paper":"https://pith.science/paper/DIQJ7VWG"},"agent_actions":{"view_html":"https://pith.science/pith/DIQJ7VWGI5ZFURS6IIJCGODEFH","download_json":"https://pith.science/pith/DIQJ7VWGI5ZFURS6IIJCGODEFH.json","view_paper":"https://pith.science/paper/DIQJ7VWG","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2502.04477&json=true","fetch_graph":"https://pith.science/api/pith-number/DIQJ7VWGI5ZFURS6IIJCGODEFH/graph.json","fetch_events":"https://pith.science/api/pith-number/DIQJ7VWGI5ZFURS6IIJCGODEFH/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/DIQJ7VWGI5ZFURS6IIJCGODEFH/action/timestamp_anchor","attest_storage":"https://pith.science/pith/DIQJ7VWGI5ZFURS6IIJCGODEFH/action/storage_attestation","attest_author":"https://pith.science/pith/DIQJ7VWGI5ZFURS6IIJCGODEFH/action/author_attestation","sign_citation":"https://pith.science/pith/DIQJ7VWGI5ZFURS6IIJCGODEFH/action/citation_signature","submit_replication":"https://pith.science/pith/DIQJ7VWGI5ZFURS6IIJCGODEFH/action/replication_record"}},"created_at":"2026-07-05T11:20:50.376838+00:00","updated_at":"2026-07-05T11:20:50.376838+00:00"}