{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:B2436R4FG5NQUNJV55BWJEPAMI","short_pith_number":"pith:B2436R4F","schema_version":"1.0","canonical_sha256":"0eb9bf4785375b0a3535ef436491e06227da209b4f011ec2dc4c4be9fec8debe","source":{"kind":"arxiv","id":"2106.01577","version":2},"attestation_state":"computed","paper":{"title":"A Provably-Efficient Model-Free Algorithm for Constrained Markov Decision Processes","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Honghao Wei, Lei Ying, Xin Liu","submitted_at":"2021-06-03T03:53:27Z","abstract_excerpt":"This paper presents the first model-free, simulator-free reinforcement learning algorithm for Constrained Markov Decision Processes (CMDPs) with sublinear regret and zero constraint violation. The algorithm is named Triple-Q because it includes three key components: a Q-function (also called action-value function) for the cumulative reward, a Q-function for the cumulative utility for the constraint, and a virtual-Queue that (over)-estimates the cumulative constraint violation. Under Triple-Q, at each step, an action is chosen based on the pseudo-Q-value that is a combination of the three \"Q\" v"},"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":"2106.01577","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2021-06-03T03:53:27Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"dfad0c68615ac40f4761e1140f9f08dd98c2c4fad36aa2865d821e950700cbda","abstract_canon_sha256":"8724d29d8d0e8c14216964879a0b4506160dc644c39ca4c6d4fef9adc01e3335"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T03:24:57.519231Z","signature_b64":"qFwdgjLC3dPYmbVQf2Zp9Ke5fXlpGMo/US6qu2xNCf7dPPy38HA1qKFBfJtrPJRodXs87uq5WLdG1nZ+UioOBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"0eb9bf4785375b0a3535ef436491e06227da209b4f011ec2dc4c4be9fec8debe","last_reissued_at":"2026-07-05T03:24:57.518782Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T03:24:57.518782Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"A Provably-Efficient Model-Free Algorithm for Constrained Markov Decision Processes","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Honghao Wei, Lei Ying, Xin Liu","submitted_at":"2021-06-03T03:53:27Z","abstract_excerpt":"This paper presents the first model-free, simulator-free reinforcement learning algorithm for Constrained Markov Decision Processes (CMDPs) with sublinear regret and zero constraint violation. The algorithm is named Triple-Q because it includes three key components: a Q-function (also called action-value function) for the cumulative reward, a Q-function for the cumulative utility for the constraint, and a virtual-Queue that (over)-estimates the cumulative constraint violation. Under Triple-Q, at each step, an action is chosen based on the pseudo-Q-value that is a combination of the three \"Q\" v"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2106.01577","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/2106.01577/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":"2106.01577","created_at":"2026-07-05T03:24:57.518840+00:00"},{"alias_kind":"arxiv_version","alias_value":"2106.01577v2","created_at":"2026-07-05T03:24:57.518840+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2106.01577","created_at":"2026-07-05T03:24:57.518840+00:00"},{"alias_kind":"pith_short_12","alias_value":"B2436R4FG5NQ","created_at":"2026-07-05T03:24:57.518840+00:00"},{"alias_kind":"pith_short_16","alias_value":"B2436R4FG5NQUNJV","created_at":"2026-07-05T03:24:57.518840+00:00"},{"alias_kind":"pith_short_8","alias_value":"B2436R4F","created_at":"2026-07-05T03:24:57.518840+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":4,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2408.16286","citing_title":"Near-Optimal Policy Identification in Robust Constrained Markov Decision Processes via Epigraph Form","ref_index":75,"is_internal_anchor":false},{"citing_arxiv_id":"2605.10519","citing_title":"Online Resource Allocation With General Constraints","ref_index":16,"is_internal_anchor":false},{"citing_arxiv_id":"2605.08290","citing_title":"Toward Optimal Regret in Robust Pricing: Decoupling Corruption and Time","ref_index":17,"is_internal_anchor":false},{"citing_arxiv_id":"2604.14243","citing_title":"Optimistic Policy Learning under Pessimistic Adversaries with Regret and Violation Guarantees","ref_index":29,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/B2436R4FG5NQUNJV55BWJEPAMI","json":"https://pith.science/pith/B2436R4FG5NQUNJV55BWJEPAMI.json","graph_json":"https://pith.science/api/pith-number/B2436R4FG5NQUNJV55BWJEPAMI/graph.json","events_json":"https://pith.science/api/pith-number/B2436R4FG5NQUNJV55BWJEPAMI/events.json","paper":"https://pith.science/paper/B2436R4F"},"agent_actions":{"view_html":"https://pith.science/pith/B2436R4FG5NQUNJV55BWJEPAMI","download_json":"https://pith.science/pith/B2436R4FG5NQUNJV55BWJEPAMI.json","view_paper":"https://pith.science/paper/B2436R4F","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2106.01577&json=true","fetch_graph":"https://pith.science/api/pith-number/B2436R4FG5NQUNJV55BWJEPAMI/graph.json","fetch_events":"https://pith.science/api/pith-number/B2436R4FG5NQUNJV55BWJEPAMI/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/B2436R4FG5NQUNJV55BWJEPAMI/action/timestamp_anchor","attest_storage":"https://pith.science/pith/B2436R4FG5NQUNJV55BWJEPAMI/action/storage_attestation","attest_author":"https://pith.science/pith/B2436R4FG5NQUNJV55BWJEPAMI/action/author_attestation","sign_citation":"https://pith.science/pith/B2436R4FG5NQUNJV55BWJEPAMI/action/citation_signature","submit_replication":"https://pith.science/pith/B2436R4FG5NQUNJV55BWJEPAMI/action/replication_record"}},"created_at":"2026-07-05T03:24:57.518840+00:00","updated_at":"2026-07-05T03:24:57.518840+00:00"}