{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2019:DMLOGIMKBQ5OQ4FVIZ4HX3JW5M","short_pith_number":"pith:DMLOGIMK","schema_version":"1.0","canonical_sha256":"1b16e3218a0c3ae870b546787bed36eb2c5dd1220fbf82434cb057bf1b184ef3","source":{"kind":"arxiv","id":"1912.06366","version":2},"attestation_state":"computed","paper":{"title":"Provably Efficient Reinforcement Learning with Aggregated States","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG","math.OC"],"primary_cat":"stat.ML","authors_text":"Benjamin Van Roy, Shi Dong, Zhengyuan Zhou","submitted_at":"2019-12-13T09:10:18Z","abstract_excerpt":"We establish that an optimistic variant of Q-learning applied to a fixed-horizon episodic Markov decision process with an aggregated state representation incurs regret $\\tilde{\\mathcal{O}}(\\sqrt{H^5 M K} + \\epsilon HK)$, where $H$ is the horizon, $M$ is the number of aggregate states, $K$ is the number of episodes, and $\\epsilon$ is the largest difference between any pair of optimal state-action values associated with a common aggregate state. Notably, this regret bound does not depend on the number of states or actions and indicates that asymptotic per-period regret is no greater than $\\epsil"},"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":"1912.06366","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"stat.ML","submitted_at":"2019-12-13T09:10:18Z","cross_cats_sorted":["cs.LG","math.OC"],"title_canon_sha256":"da8a3acef53a90a5571056e6bdbf3a6af3bcaca6642aefa5820bd8c054f8c016","abstract_canon_sha256":"29ac0d533a859c398531fe4b90d189a8c658021cab32d4bbe00b611e90b637c0"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T00:42:22.259567Z","signature_b64":"tbrn4m674eJIXIucHpuac3socMSJpgxtqs2qNdN26QCZjP5477evbda8UNwTWef8csjSdZ1Z4UnDAVegB1mxDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"1b16e3218a0c3ae870b546787bed36eb2c5dd1220fbf82434cb057bf1b184ef3","last_reissued_at":"2026-07-05T00:42:22.259055Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T00:42:22.259055Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Provably Efficient Reinforcement Learning with Aggregated States","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG","math.OC"],"primary_cat":"stat.ML","authors_text":"Benjamin Van Roy, Shi Dong, Zhengyuan Zhou","submitted_at":"2019-12-13T09:10:18Z","abstract_excerpt":"We establish that an optimistic variant of Q-learning applied to a fixed-horizon episodic Markov decision process with an aggregated state representation incurs regret $\\tilde{\\mathcal{O}}(\\sqrt{H^5 M K} + \\epsilon HK)$, where $H$ is the horizon, $M$ is the number of aggregate states, $K$ is the number of episodes, and $\\epsilon$ is the largest difference between any pair of optimal state-action values associated with a common aggregate state. Notably, this regret bound does not depend on the number of states or actions and indicates that asymptotic per-period regret is no greater than $\\epsil"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1912.06366","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/1912.06366/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":"1912.06366","created_at":"2026-07-05T00:42:22.259120+00:00"},{"alias_kind":"arxiv_version","alias_value":"1912.06366v2","created_at":"2026-07-05T00:42:22.259120+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1912.06366","created_at":"2026-07-05T00:42:22.259120+00:00"},{"alias_kind":"pith_short_12","alias_value":"DMLOGIMKBQ5O","created_at":"2026-07-05T00:42:22.259120+00:00"},{"alias_kind":"pith_short_16","alias_value":"DMLOGIMKBQ5OQ4FV","created_at":"2026-07-05T00:42:22.259120+00:00"},{"alias_kind":"pith_short_8","alias_value":"DMLOGIMK","created_at":"2026-07-05T00:42:22.259120+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.28276","citing_title":"Commit to the Bit: Reactive Reinforcement Learning Done Right","ref_index":1,"is_internal_anchor":false},{"citing_arxiv_id":"2605.00654","citing_title":"Reinforcement Learning with Markov Risk Measures and Multipattern Risk Approximation","ref_index":21,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/DMLOGIMKBQ5OQ4FVIZ4HX3JW5M","json":"https://pith.science/pith/DMLOGIMKBQ5OQ4FVIZ4HX3JW5M.json","graph_json":"https://pith.science/api/pith-number/DMLOGIMKBQ5OQ4FVIZ4HX3JW5M/graph.json","events_json":"https://pith.science/api/pith-number/DMLOGIMKBQ5OQ4FVIZ4HX3JW5M/events.json","paper":"https://pith.science/paper/DMLOGIMK"},"agent_actions":{"view_html":"https://pith.science/pith/DMLOGIMKBQ5OQ4FVIZ4HX3JW5M","download_json":"https://pith.science/pith/DMLOGIMKBQ5OQ4FVIZ4HX3JW5M.json","view_paper":"https://pith.science/paper/DMLOGIMK","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1912.06366&json=true","fetch_graph":"https://pith.science/api/pith-number/DMLOGIMKBQ5OQ4FVIZ4HX3JW5M/graph.json","fetch_events":"https://pith.science/api/pith-number/DMLOGIMKBQ5OQ4FVIZ4HX3JW5M/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/DMLOGIMKBQ5OQ4FVIZ4HX3JW5M/action/timestamp_anchor","attest_storage":"https://pith.science/pith/DMLOGIMKBQ5OQ4FVIZ4HX3JW5M/action/storage_attestation","attest_author":"https://pith.science/pith/DMLOGIMKBQ5OQ4FVIZ4HX3JW5M/action/author_attestation","sign_citation":"https://pith.science/pith/DMLOGIMKBQ5OQ4FVIZ4HX3JW5M/action/citation_signature","submit_replication":"https://pith.science/pith/DMLOGIMKBQ5OQ4FVIZ4HX3JW5M/action/replication_record"}},"created_at":"2026-07-05T00:42:22.259120+00:00","updated_at":"2026-07-05T00:42:22.259120+00:00"}