{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2019:BRACNNQIGXMFCLTVB4ISOS7EMM","short_pith_number":"pith:BRACNNQI","schema_version":"1.0","canonical_sha256":"0c4026b60835d8512e750f11274be4630b89f1387f9fe617a9c41692ab6efd4d","source":{"kind":"arxiv","id":"1905.06265","version":2},"attestation_state":"computed","paper":{"title":"Stochastic approximation with cone-contractive operators: Sharp $\\ell_\\infty$-bounds for $Q$-learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["math.OC","stat.ML"],"primary_cat":"cs.LG","authors_text":"Martin J. Wainwright","submitted_at":"2019-05-15T16:06:30Z","abstract_excerpt":"Motivated by the study of $Q$-learning algorithms in reinforcement learning, we study a class of stochastic approximation procedures based on operators that satisfy monotonicity and quasi-contractivity conditions with respect to an underlying cone. We prove a general sandwich relation on the iterate error at each time, and use it to derive non-asymptotic bounds on the error in terms of a cone-induced gauge norm. These results are derived within a deterministic framework, requiring no assumptions on the noise. We illustrate these general bounds in application to synchronous $Q$-learning for dis"},"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":"1905.06265","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2019-05-15T16:06:30Z","cross_cats_sorted":["math.OC","stat.ML"],"title_canon_sha256":"fcc03765db7c3ce5ae0d7bd3ef59159556c6e995273be4c6dbf0d97e3ecaa4c0","abstract_canon_sha256":"19f6b7c89245aa485c1021726f02c2e8765b70652c7e6547468c86613852d203"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-17T23:42:41.661064Z","signature_b64":"cDPe7CSwlfgcX3xQUnF1hX4HDdfPsAsEb99vZOSeFqFKf1k2STGSQ/rC6xHQUTQJL38UdC2FL9Iim5OefU/tDw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"0c4026b60835d8512e750f11274be4630b89f1387f9fe617a9c41692ab6efd4d","last_reissued_at":"2026-05-17T23:42:41.660241Z","signature_status":"signed_v1","first_computed_at":"2026-05-17T23:42:41.660241Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Stochastic approximation with cone-contractive operators: Sharp $\\ell_\\infty$-bounds for $Q$-learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["math.OC","stat.ML"],"primary_cat":"cs.LG","authors_text":"Martin J. Wainwright","submitted_at":"2019-05-15T16:06:30Z","abstract_excerpt":"Motivated by the study of $Q$-learning algorithms in reinforcement learning, we study a class of stochastic approximation procedures based on operators that satisfy monotonicity and quasi-contractivity conditions with respect to an underlying cone. We prove a general sandwich relation on the iterate error at each time, and use it to derive non-asymptotic bounds on the error in terms of a cone-induced gauge norm. These results are derived within a deterministic framework, requiring no assumptions on the noise. We illustrate these general bounds in application to synchronous $Q$-learning for dis"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1905.06265","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":""},"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":"1905.06265","created_at":"2026-05-17T23:42:41.660403+00:00"},{"alias_kind":"arxiv_version","alias_value":"1905.06265v2","created_at":"2026-05-17T23:42:41.660403+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1905.06265","created_at":"2026-05-17T23:42:41.660403+00:00"},{"alias_kind":"pith_short_12","alias_value":"BRACNNQIGXMF","created_at":"2026-05-18T12:33:12.712433+00:00"},{"alias_kind":"pith_short_16","alias_value":"BRACNNQIGXMFCLTV","created_at":"2026-05-18T12:33:12.712433+00:00"},{"alias_kind":"pith_short_8","alias_value":"BRACNNQI","created_at":"2026-05-18T12:33:12.712433+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":8,"internal_anchor_count":6,"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":12,"is_internal_anchor":true},{"citing_arxiv_id":"2504.18743","citing_title":"From Set Convergence to Pointwise Convergence: Finite-Time Guarantees for Average-Reward Q-Learning with Adaptive Stepsizes","ref_index":74,"is_internal_anchor":true},{"citing_arxiv_id":"2509.08933","citing_title":"Corruption-Tolerant Asynchronous Q-Learning with Near-Optimal Rates","ref_index":58,"is_internal_anchor":true},{"citing_arxiv_id":"2605.16103","citing_title":"Sign-Separated Finite-Time Error Analysis of Q-Learning","ref_index":29,"is_internal_anchor":true},{"citing_arxiv_id":"2509.18964","citing_title":"Central Limit Theorems for Asynchronous Averaged Q-Learning","ref_index":11,"is_internal_anchor":true},{"citing_arxiv_id":"2510.16132","citing_title":"A Minimal-Assumption Analysis of Q-Learning with Time-Varying Policies","ref_index":17,"is_internal_anchor":true},{"citing_arxiv_id":"2604.19569","citing_title":"Lyapunov-Certified Direct Switching Theory for Q-Learning","ref_index":29,"is_internal_anchor":false},{"citing_arxiv_id":"2604.07323","citing_title":"Gaussian Approximation for Asynchronous Q-learning","ref_index":48,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/BRACNNQIGXMFCLTVB4ISOS7EMM","json":"https://pith.science/pith/BRACNNQIGXMFCLTVB4ISOS7EMM.json","graph_json":"https://pith.science/api/pith-number/BRACNNQIGXMFCLTVB4ISOS7EMM/graph.json","events_json":"https://pith.science/api/pith-number/BRACNNQIGXMFCLTVB4ISOS7EMM/events.json","paper":"https://pith.science/paper/BRACNNQI"},"agent_actions":{"view_html":"https://pith.science/pith/BRACNNQIGXMFCLTVB4ISOS7EMM","download_json":"https://pith.science/pith/BRACNNQIGXMFCLTVB4ISOS7EMM.json","view_paper":"https://pith.science/paper/BRACNNQI","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1905.06265&json=true","fetch_graph":"https://pith.science/api/pith-number/BRACNNQIGXMFCLTVB4ISOS7EMM/graph.json","fetch_events":"https://pith.science/api/pith-number/BRACNNQIGXMFCLTVB4ISOS7EMM/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/BRACNNQIGXMFCLTVB4ISOS7EMM/action/timestamp_anchor","attest_storage":"https://pith.science/pith/BRACNNQIGXMFCLTVB4ISOS7EMM/action/storage_attestation","attest_author":"https://pith.science/pith/BRACNNQIGXMFCLTVB4ISOS7EMM/action/author_attestation","sign_citation":"https://pith.science/pith/BRACNNQIGXMFCLTVB4ISOS7EMM/action/citation_signature","submit_replication":"https://pith.science/pith/BRACNNQIGXMFCLTVB4ISOS7EMM/action/replication_record"}},"created_at":"2026-05-17T23:42:41.660403+00:00","updated_at":"2026-05-17T23:42:41.660403+00:00"}