{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:OFGEU7XIBYMWSRX5GWEGCZGENH","short_pith_number":"pith:OFGEU7XI","schema_version":"1.0","canonical_sha256":"714c4a7ee80e196946fd35886164c469fbaa23d7847e36f9163d8e3a1eb77e5d","source":{"kind":"arxiv","id":"2408.10381","version":1},"attestation_state":"computed","paper":{"title":"Efficient Reinforcement Learning in Probabilistic Reward Machines","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"stat.ML","authors_text":"Xiaofeng Lin, Xuezhou Zhang","submitted_at":"2024-08-19T19:51:53Z","abstract_excerpt":"In this paper, we study reinforcement learning in Markov Decision Processes with Probabilistic Reward Machines (PRMs), a form of non-Markovian reward commonly found in robotics tasks. We design an algorithm for PRMs that achieves a regret bound of $\\widetilde{O}(\\sqrt{HOAT} + H^2O^2A^{3/2} + H\\sqrt{T})$, where $H$ is the time horizon, $O$ is the number of observations, $A$ is the number of actions, and $T$ is the number of time-steps. This result improves over the best-known bound, $\\widetilde{O}(H\\sqrt{OAT})$ of \\citet{pmlr-v206-bourel23a} for MDPs with Deterministic Reward Machines (DRMs), a"},"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":"2408.10381","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"stat.ML","submitted_at":"2024-08-19T19:51:53Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"4c6ffbb4979785e7db2741e801334e0a3faf5d0b62e69edb79d6caf9953ccb75","abstract_canon_sha256":"78dc86b7ad7df83b1722bbd2dec670f0a6f6fc446f9131fa853941cd55cbb425"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:57:11.078639Z","signature_b64":"7t7slbQerhkYYxFtgE6XWfuQ5DiWqoutd7JFx94FoiYgo6WQiwQun3NCHcGmpfx0Cngofyofin3gXIq4qu/ZAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"714c4a7ee80e196946fd35886164c469fbaa23d7847e36f9163d8e3a1eb77e5d","last_reissued_at":"2026-07-05T08:57:11.078255Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:57:11.078255Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Efficient Reinforcement Learning in Probabilistic Reward Machines","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"stat.ML","authors_text":"Xiaofeng Lin, Xuezhou Zhang","submitted_at":"2024-08-19T19:51:53Z","abstract_excerpt":"In this paper, we study reinforcement learning in Markov Decision Processes with Probabilistic Reward Machines (PRMs), a form of non-Markovian reward commonly found in robotics tasks. We design an algorithm for PRMs that achieves a regret bound of $\\widetilde{O}(\\sqrt{HOAT} + H^2O^2A^{3/2} + H\\sqrt{T})$, where $H$ is the time horizon, $O$ is the number of observations, $A$ is the number of actions, and $T$ is the number of time-steps. This result improves over the best-known bound, $\\widetilde{O}(H\\sqrt{OAT})$ of \\citet{pmlr-v206-bourel23a} for MDPs with Deterministic Reward Machines (DRMs), a"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2408.10381","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/2408.10381/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":"2408.10381","created_at":"2026-07-05T08:57:11.078306+00:00"},{"alias_kind":"arxiv_version","alias_value":"2408.10381v1","created_at":"2026-07-05T08:57:11.078306+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2408.10381","created_at":"2026-07-05T08:57:11.078306+00:00"},{"alias_kind":"pith_short_12","alias_value":"OFGEU7XIBYMW","created_at":"2026-07-05T08:57:11.078306+00:00"},{"alias_kind":"pith_short_16","alias_value":"OFGEU7XIBYMWSRX5","created_at":"2026-07-05T08:57:11.078306+00:00"},{"alias_kind":"pith_short_8","alias_value":"OFGEU7XI","created_at":"2026-07-05T08:57:11.078306+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2508.14093","citing_title":"Physics-Informed Reward Machines","ref_index":2006,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/OFGEU7XIBYMWSRX5GWEGCZGENH","json":"https://pith.science/pith/OFGEU7XIBYMWSRX5GWEGCZGENH.json","graph_json":"https://pith.science/api/pith-number/OFGEU7XIBYMWSRX5GWEGCZGENH/graph.json","events_json":"https://pith.science/api/pith-number/OFGEU7XIBYMWSRX5GWEGCZGENH/events.json","paper":"https://pith.science/paper/OFGEU7XI"},"agent_actions":{"view_html":"https://pith.science/pith/OFGEU7XIBYMWSRX5GWEGCZGENH","download_json":"https://pith.science/pith/OFGEU7XIBYMWSRX5GWEGCZGENH.json","view_paper":"https://pith.science/paper/OFGEU7XI","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2408.10381&json=true","fetch_graph":"https://pith.science/api/pith-number/OFGEU7XIBYMWSRX5GWEGCZGENH/graph.json","fetch_events":"https://pith.science/api/pith-number/OFGEU7XIBYMWSRX5GWEGCZGENH/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/OFGEU7XIBYMWSRX5GWEGCZGENH/action/timestamp_anchor","attest_storage":"https://pith.science/pith/OFGEU7XIBYMWSRX5GWEGCZGENH/action/storage_attestation","attest_author":"https://pith.science/pith/OFGEU7XIBYMWSRX5GWEGCZGENH/action/author_attestation","sign_citation":"https://pith.science/pith/OFGEU7XIBYMWSRX5GWEGCZGENH/action/citation_signature","submit_replication":"https://pith.science/pith/OFGEU7XIBYMWSRX5GWEGCZGENH/action/replication_record"}},"created_at":"2026-07-05T08:57:11.078306+00:00","updated_at":"2026-07-05T08:57:11.078306+00:00"}