{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2013:UKOXY56QHBJEVEZY63XTLGS7OR","short_pith_number":"pith:UKOXY56Q","schema_version":"1.0","canonical_sha256":"a29d7c77d038524a9338f6ef359a5f746306b661bbc67240f3b9fdbbd46d9ff4","source":{"kind":"arxiv","id":"1301.2315","version":1},"attestation_state":"computed","paper":{"title":"The Optimal Reward Baseline for Gradient-Based Reinforcement Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","stat.ML"],"primary_cat":"cs.LG","authors_text":"Lex Weaver, Nigel Tao","submitted_at":"2013-01-10T16:26:53Z","abstract_excerpt":"There exist a number of reinforcement learning algorithms which learnby climbing the gradient of expected reward.  Their long-runconvergence has been proved, even in partially observableenvironments with non-deterministic actions, and without the need fora system model.  However, the variance of the gradient estimator hasbeen found to be a significant practical problem.  Recent approacheshave discounted future rewards, introducing a bias-variance trade-offinto the gradient estimate.  We incorporate a reward baseline into thelearning system, and show that it affects variance without introducing"},"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":"1301.2315","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2013-01-10T16:26:53Z","cross_cats_sorted":["cs.AI","stat.ML"],"title_canon_sha256":"0512690abe75423e8d2b65f491496bad633cede55d51dda2379cb0e94f068227","abstract_canon_sha256":"51205d32ac70a94d8142adcd291dcc7e6f16f73e7e0c592c22384d976eb787c7"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-18T03:36:45.341659Z","signature_b64":"V8NwyKb+N9F+mAtgY6pkwprGNm4MAFj1oqA13uNdwvECXfTth0qZK/Pso0+4pbQOJFJysZMlNabIID40vXt2Bg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"a29d7c77d038524a9338f6ef359a5f746306b661bbc67240f3b9fdbbd46d9ff4","last_reissued_at":"2026-05-18T03:36:45.340956Z","signature_status":"signed_v1","first_computed_at":"2026-05-18T03:36:45.340956Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"The Optimal Reward Baseline for Gradient-Based Reinforcement Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","stat.ML"],"primary_cat":"cs.LG","authors_text":"Lex Weaver, Nigel Tao","submitted_at":"2013-01-10T16:26:53Z","abstract_excerpt":"There exist a number of reinforcement learning algorithms which learnby climbing the gradient of expected reward.  Their long-runconvergence has been proved, even in partially observableenvironments with non-deterministic actions, and without the need fora system model.  However, the variance of the gradient estimator hasbeen found to be a significant practical problem.  Recent approacheshave discounted future rewards, introducing a bias-variance trade-offinto the gradient estimate.  We incorporate a reward baseline into thelearning system, and show that it affects variance without introducing"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1301.2315","kind":"arxiv","version":1},"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":"1301.2315","created_at":"2026-05-18T03:36:45.341072+00:00"},{"alias_kind":"arxiv_version","alias_value":"1301.2315v1","created_at":"2026-05-18T03:36:45.341072+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1301.2315","created_at":"2026-05-18T03:36:45.341072+00:00"},{"alias_kind":"pith_short_12","alias_value":"UKOXY56QHBJE","created_at":"2026-05-18T12:28:02.375192+00:00"},{"alias_kind":"pith_short_16","alias_value":"UKOXY56QHBJEVEZY","created_at":"2026-05-18T12:28:02.375192+00:00"},{"alias_kind":"pith_short_8","alias_value":"UKOXY56Q","created_at":"2026-05-18T12:28:02.375192+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":2,"sample":[{"citing_arxiv_id":"2606.06080","citing_title":"On Advantage Estimates for Max@K Policy Gradients","ref_index":56,"is_internal_anchor":true},{"citing_arxiv_id":"2410.09355","citing_title":"On Divergence Measures for Training GFlowNets","ref_index":94,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/UKOXY56QHBJEVEZY63XTLGS7OR","json":"https://pith.science/pith/UKOXY56QHBJEVEZY63XTLGS7OR.json","graph_json":"https://pith.science/api/pith-number/UKOXY56QHBJEVEZY63XTLGS7OR/graph.json","events_json":"https://pith.science/api/pith-number/UKOXY56QHBJEVEZY63XTLGS7OR/events.json","paper":"https://pith.science/paper/UKOXY56Q"},"agent_actions":{"view_html":"https://pith.science/pith/UKOXY56QHBJEVEZY63XTLGS7OR","download_json":"https://pith.science/pith/UKOXY56QHBJEVEZY63XTLGS7OR.json","view_paper":"https://pith.science/paper/UKOXY56Q","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1301.2315&json=true","fetch_graph":"https://pith.science/api/pith-number/UKOXY56QHBJEVEZY63XTLGS7OR/graph.json","fetch_events":"https://pith.science/api/pith-number/UKOXY56QHBJEVEZY63XTLGS7OR/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/UKOXY56QHBJEVEZY63XTLGS7OR/action/timestamp_anchor","attest_storage":"https://pith.science/pith/UKOXY56QHBJEVEZY63XTLGS7OR/action/storage_attestation","attest_author":"https://pith.science/pith/UKOXY56QHBJEVEZY63XTLGS7OR/action/author_attestation","sign_citation":"https://pith.science/pith/UKOXY56QHBJEVEZY63XTLGS7OR/action/citation_signature","submit_replication":"https://pith.science/pith/UKOXY56QHBJEVEZY63XTLGS7OR/action/replication_record"}},"created_at":"2026-05-18T03:36:45.341072+00:00","updated_at":"2026-05-18T03:36:45.341072+00:00"}