{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:QKDDNVFS4CZ2V2H57RNBCOQG6Q","short_pith_number":"pith:QKDDNVFS","schema_version":"1.0","canonical_sha256":"828636d4b2e0b3aae8fdfc5a113a06f40718532e13a591f2b15720053abc011c","source":{"kind":"arxiv","id":"2102.01567","version":4},"attestation_state":"computed","paper":{"title":"A Lyapunov Theory for Finite-Sample Guarantees of Asynchronous Q-Learning and TD-Learning Variants","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["math.OC","stat.ML"],"primary_cat":"cs.LG","authors_text":"Karthikeyan Shanmugam, Sanjay Shakkottai, Siva Theja Maguluri, Zaiwei Chen","submitted_at":"2021-02-02T15:48:19Z","abstract_excerpt":"This paper develops an unified framework to study finite-sample convergence guarantees of a large class of value-based asynchronous reinforcement learning (RL) algorithms. We do this by first reformulating the RL algorithms as \\textit{Markovian Stochastic Approximation} (SA) algorithms to solve fixed-point equations. We then develop a Lyapunov analysis and derive mean-square error bounds on the convergence of the Markovian SA. Based on this result, we establish finite-sample mean-square convergence bounds for asynchronous RL algorithms such as $Q$-learning, $n$-step TD, TD$(\\lambda)$, and off-"},"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":"2102.01567","kind":"arxiv","version":4},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2021-02-02T15:48:19Z","cross_cats_sorted":["math.OC","stat.ML"],"title_canon_sha256":"5981ff223a079d70b12affaf25090a9d9c8158fd34d1b9246cc5e8b48fd872f4","abstract_canon_sha256":"73f96d70148586e6c87c70bc5463a5c39361e8c487c37dae93c7cfc1474903ae"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T06:47:10.476102Z","signature_b64":"7iqYsIqOnq57P1c5AJfHYNgO+OdK4S4qc95DAHnqGorteu5Bte+5nml5iG+AzO9cyhuOA7oluH+3NZfU7jhoDg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"828636d4b2e0b3aae8fdfc5a113a06f40718532e13a591f2b15720053abc011c","last_reissued_at":"2026-07-05T06:47:10.475601Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T06:47:10.475601Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"A Lyapunov Theory for Finite-Sample Guarantees of Asynchronous Q-Learning and TD-Learning Variants","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["math.OC","stat.ML"],"primary_cat":"cs.LG","authors_text":"Karthikeyan Shanmugam, Sanjay Shakkottai, Siva Theja Maguluri, Zaiwei Chen","submitted_at":"2021-02-02T15:48:19Z","abstract_excerpt":"This paper develops an unified framework to study finite-sample convergence guarantees of a large class of value-based asynchronous reinforcement learning (RL) algorithms. We do this by first reformulating the RL algorithms as \\textit{Markovian Stochastic Approximation} (SA) algorithms to solve fixed-point equations. We then develop a Lyapunov analysis and derive mean-square error bounds on the convergence of the Markovian SA. Based on this result, we establish finite-sample mean-square convergence bounds for asynchronous RL algorithms such as $Q$-learning, $n$-step TD, TD$(\\lambda)$, and off-"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2102.01567","kind":"arxiv","version":4},"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/2102.01567/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":"2102.01567","created_at":"2026-07-05T06:47:10.475659+00:00"},{"alias_kind":"arxiv_version","alias_value":"2102.01567v4","created_at":"2026-07-05T06:47:10.475659+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2102.01567","created_at":"2026-07-05T06:47:10.475659+00:00"},{"alias_kind":"pith_short_12","alias_value":"QKDDNVFS4CZ2","created_at":"2026-07-05T06:47:10.475659+00:00"},{"alias_kind":"pith_short_16","alias_value":"QKDDNVFS4CZ2V2H5","created_at":"2026-07-05T06:47:10.475659+00:00"},{"alias_kind":"pith_short_8","alias_value":"QKDDNVFS","created_at":"2026-07-05T06:47:10.475659+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":9,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2404.14442","citing_title":"Toward a Unified Lyapunov-Certified ODE Convergence Analysis of Smooth Q-Learning with p-Norms","ref_index":15,"is_internal_anchor":false},{"citing_arxiv_id":"2605.20999","citing_title":"Concentration of General Stochastic Approximation Under Heavy-Tailed Markovian Noise","ref_index":206,"is_internal_anchor":false},{"citing_arxiv_id":"2605.11021","citing_title":"A Switching System Theory of Q-Learning with Linear Function Approximation","ref_index":6,"is_internal_anchor":false},{"citing_arxiv_id":"2605.16103","citing_title":"Sign-Separated Asymmetric Finite-Time Error Analysis of Q-Learning","ref_index":5,"is_internal_anchor":false},{"citing_arxiv_id":"2509.18964","citing_title":"Central Limit Theorems for Asynchronous Averaged Q-Learning","ref_index":3,"is_internal_anchor":false},{"citing_arxiv_id":"2605.11021","citing_title":"A Switching System Theory of Q-Learning with Linear Function Approximation","ref_index":6,"is_internal_anchor":false},{"citing_arxiv_id":"2604.19569","citing_title":"Switching Theory for Q-Learning","ref_index":5,"is_internal_anchor":false},{"citing_arxiv_id":"2605.08053","citing_title":"Reinforcement Learning for Exponential Utility: Algorithms and Convergence in Discounted MDPs","ref_index":45,"is_internal_anchor":false},{"citing_arxiv_id":"2604.04394","citing_title":"Finite-Time Analysis of Q-Value Iteration for General-Sum Stackelberg Games","ref_index":4,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/QKDDNVFS4CZ2V2H57RNBCOQG6Q","json":"https://pith.science/pith/QKDDNVFS4CZ2V2H57RNBCOQG6Q.json","graph_json":"https://pith.science/api/pith-number/QKDDNVFS4CZ2V2H57RNBCOQG6Q/graph.json","events_json":"https://pith.science/api/pith-number/QKDDNVFS4CZ2V2H57RNBCOQG6Q/events.json","paper":"https://pith.science/paper/QKDDNVFS"},"agent_actions":{"view_html":"https://pith.science/pith/QKDDNVFS4CZ2V2H57RNBCOQG6Q","download_json":"https://pith.science/pith/QKDDNVFS4CZ2V2H57RNBCOQG6Q.json","view_paper":"https://pith.science/paper/QKDDNVFS","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2102.01567&json=true","fetch_graph":"https://pith.science/api/pith-number/QKDDNVFS4CZ2V2H57RNBCOQG6Q/graph.json","fetch_events":"https://pith.science/api/pith-number/QKDDNVFS4CZ2V2H57RNBCOQG6Q/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/QKDDNVFS4CZ2V2H57RNBCOQG6Q/action/timestamp_anchor","attest_storage":"https://pith.science/pith/QKDDNVFS4CZ2V2H57RNBCOQG6Q/action/storage_attestation","attest_author":"https://pith.science/pith/QKDDNVFS4CZ2V2H57RNBCOQG6Q/action/author_attestation","sign_citation":"https://pith.science/pith/QKDDNVFS4CZ2V2H57RNBCOQG6Q/action/citation_signature","submit_replication":"https://pith.science/pith/QKDDNVFS4CZ2V2H57RNBCOQG6Q/action/replication_record"}},"created_at":"2026-07-05T06:47:10.475659+00:00","updated_at":"2026-07-05T06:47:10.475659+00:00"}