{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:Y4HIXO76SO2SULYI2LKGCKTZLZ","short_pith_number":"pith:Y4HIXO76","schema_version":"1.0","canonical_sha256":"c70e8bbbfe93b52a2f08d2d4612a795e67230046ee447d2c1328b743f7d6ab2a","source":{"kind":"arxiv","id":"2501.19254","version":4},"attestation_state":"computed","paper":{"title":"Linear $Q$-Learning Does Not Diverge in $L^2$: Convergence Rates to a Bounded Set","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","stat.ML"],"primary_cat":"cs.LG","authors_text":"Shangtong Zhang, Xinyu Liu, Zixuan Xie","submitted_at":"2025-01-31T16:10:50Z","abstract_excerpt":"$Q$-learning is one of the most fundamental reinforcement learning algorithms. It is widely believed that $Q$-learning with linear function approximation (i.e., linear $Q$-learning) suffers from possible divergence until the recent work Meyn (2024) which establishes the ultimate almost sure boundedness of the iterates of linear $Q$-learning. Building on this success, this paper further establishes the first $L^2$ convergence rate of linear $Q$-learning iterates (to a bounded set). Similar to Meyn (2024), we do not make any modification to the original linear $Q$-learning algorithm, do not make"},"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":"2501.19254","kind":"arxiv","version":4},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-01-31T16:10:50Z","cross_cats_sorted":["cs.AI","stat.ML"],"title_canon_sha256":"b0991523154d4ceddb3eaae02a6874f509bd9e067ee967bae11b6909760db017","abstract_canon_sha256":"ff9c1a2d3dd66910d084d463cd964bbffaca54b88a248a1876e2405e27204fa5"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:10:34.338746Z","signature_b64":"vDI9LIVEzmf7rgCTn+yw5bvzJ5rTL9YSc0iENa0GYsYPuqZc7B/f+TMvOjQDJIPFhWVeRcGKjpFDKgjJA806Cw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"c70e8bbbfe93b52a2f08d2d4612a795e67230046ee447d2c1328b743f7d6ab2a","last_reissued_at":"2026-07-05T11:10:34.338262Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:10:34.338262Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Linear $Q$-Learning Does Not Diverge in $L^2$: Convergence Rates to a Bounded Set","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","stat.ML"],"primary_cat":"cs.LG","authors_text":"Shangtong Zhang, Xinyu Liu, Zixuan Xie","submitted_at":"2025-01-31T16:10:50Z","abstract_excerpt":"$Q$-learning is one of the most fundamental reinforcement learning algorithms. It is widely believed that $Q$-learning with linear function approximation (i.e., linear $Q$-learning) suffers from possible divergence until the recent work Meyn (2024) which establishes the ultimate almost sure boundedness of the iterates of linear $Q$-learning. Building on this success, this paper further establishes the first $L^2$ convergence rate of linear $Q$-learning iterates (to a bounded set). Similar to Meyn (2024), we do not make any modification to the original linear $Q$-learning algorithm, do not make"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2501.19254","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/2501.19254/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":"2501.19254","created_at":"2026-07-05T11:10:34.338322+00:00"},{"alias_kind":"arxiv_version","alias_value":"2501.19254v4","created_at":"2026-07-05T11:10:34.338322+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2501.19254","created_at":"2026-07-05T11:10:34.338322+00:00"},{"alias_kind":"pith_short_12","alias_value":"Y4HIXO76SO2S","created_at":"2026-07-05T11:10:34.338322+00:00"},{"alias_kind":"pith_short_16","alias_value":"Y4HIXO76SO2SULYI","created_at":"2026-07-05T11:10:34.338322+00:00"},{"alias_kind":"pith_short_8","alias_value":"Y4HIXO76","created_at":"2026-07-05T11:10:34.338322+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":3,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.11021","citing_title":"A Switching System Theory of Q-Learning with Linear Function Approximation","ref_index":24,"is_internal_anchor":false},{"citing_arxiv_id":"2510.16132","citing_title":"A Minimal-Assumption Analysis of Q-Learning with Time-Varying Policies","ref_index":25,"is_internal_anchor":false},{"citing_arxiv_id":"2605.11021","citing_title":"A Switching System Theory of Q-Learning with Linear Function Approximation","ref_index":24,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/Y4HIXO76SO2SULYI2LKGCKTZLZ","json":"https://pith.science/pith/Y4HIXO76SO2SULYI2LKGCKTZLZ.json","graph_json":"https://pith.science/api/pith-number/Y4HIXO76SO2SULYI2LKGCKTZLZ/graph.json","events_json":"https://pith.science/api/pith-number/Y4HIXO76SO2SULYI2LKGCKTZLZ/events.json","paper":"https://pith.science/paper/Y4HIXO76"},"agent_actions":{"view_html":"https://pith.science/pith/Y4HIXO76SO2SULYI2LKGCKTZLZ","download_json":"https://pith.science/pith/Y4HIXO76SO2SULYI2LKGCKTZLZ.json","view_paper":"https://pith.science/paper/Y4HIXO76","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2501.19254&json=true","fetch_graph":"https://pith.science/api/pith-number/Y4HIXO76SO2SULYI2LKGCKTZLZ/graph.json","fetch_events":"https://pith.science/api/pith-number/Y4HIXO76SO2SULYI2LKGCKTZLZ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/Y4HIXO76SO2SULYI2LKGCKTZLZ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/Y4HIXO76SO2SULYI2LKGCKTZLZ/action/storage_attestation","attest_author":"https://pith.science/pith/Y4HIXO76SO2SULYI2LKGCKTZLZ/action/author_attestation","sign_citation":"https://pith.science/pith/Y4HIXO76SO2SULYI2LKGCKTZLZ/action/citation_signature","submit_replication":"https://pith.science/pith/Y4HIXO76SO2SULYI2LKGCKTZLZ/action/replication_record"}},"created_at":"2026-07-05T11:10:34.338322+00:00","updated_at":"2026-07-05T11:10:34.338322+00:00"}