{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2019:2ISVUUXERZSSO4TPRGY5WHTCDA","short_pith_number":"pith:2ISVUUXE","schema_version":"1.0","canonical_sha256":"d2255a52e48e6527726f89b1db1e621820000d6523f6404e7c42563c97ab9a8d","source":{"kind":"arxiv","id":"1906.00949","version":2},"attestation_state":"computed","paper":{"title":"Stabilizing Off-Policy Q-Learning via Bootstrapping Error Reduction","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Aviral Kumar, George Tucker, Justin Fu, Sergey Levine","submitted_at":"2019-06-03T17:59:10Z","abstract_excerpt":"Off-policy reinforcement learning aims to leverage experience collected from prior policies for sample-efficient learning. However, in practice, commonly used off-policy approximate dynamic programming methods based on Q-learning and actor-critic methods are highly sensitive to the data distribution, and can make only limited progress without collecting additional on-policy data. As a step towards more robust off-policy algorithms, we study the setting where the off-policy experience is fixed and there is no further interaction with the environment. We identify bootstrapping error as a key sou"},"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":"1906.00949","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2019-06-03T17:59:10Z","cross_cats_sorted":["stat.ML"],"title_canon_sha256":"e2e67f948f0c00e2b41e6e20194a4fe52bd859bfea3b16f19b555a9ccfc91b97","abstract_canon_sha256":"6f84ed6861abc2e9dcd943e4a4a92fa28ae0496dae5094ca96a2b680fa36b280"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T00:21:34.798089Z","signature_b64":"sp449vgeMD7WgYlvcCO9LBTmCgHYEdxId1h2UZrpJS2LUcYIRbwg/qOcZd2NaffKz4DNImC9aUp4uil0U0dvCg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"d2255a52e48e6527726f89b1db1e621820000d6523f6404e7c42563c97ab9a8d","last_reissued_at":"2026-07-05T00:21:34.797567Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T00:21:34.797567Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Stabilizing Off-Policy Q-Learning via Bootstrapping Error Reduction","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Aviral Kumar, George Tucker, Justin Fu, Sergey Levine","submitted_at":"2019-06-03T17:59:10Z","abstract_excerpt":"Off-policy reinforcement learning aims to leverage experience collected from prior policies for sample-efficient learning. However, in practice, commonly used off-policy approximate dynamic programming methods based on Q-learning and actor-critic methods are highly sensitive to the data distribution, and can make only limited progress without collecting additional on-policy data. As a step towards more robust off-policy algorithms, we study the setting where the off-policy experience is fixed and there is no further interaction with the environment. We identify bootstrapping error as a key sou"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1906.00949","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/1906.00949/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":"1906.00949","created_at":"2026-07-05T00:21:34.797628+00:00"},{"alias_kind":"arxiv_version","alias_value":"1906.00949v2","created_at":"2026-07-05T00:21:34.797628+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1906.00949","created_at":"2026-07-05T00:21:34.797628+00:00"},{"alias_kind":"pith_short_12","alias_value":"2ISVUUXERZSS","created_at":"2026-07-05T00:21:34.797628+00:00"},{"alias_kind":"pith_short_16","alias_value":"2ISVUUXERZSSO4TP","created_at":"2026-07-05T00:21:34.797628+00:00"},{"alias_kind":"pith_short_8","alias_value":"2ISVUUXE","created_at":"2026-07-05T00:21:34.797628+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":11,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.00350","citing_title":"Drift Q-Learning","ref_index":90,"is_internal_anchor":false},{"citing_arxiv_id":"2605.11151","citing_title":"RankQ: Offline-to-Online Reinforcement Learning via Self-Supervised Action Ranking","ref_index":10,"is_internal_anchor":false},{"citing_arxiv_id":"2605.18320","citing_title":"ISEP: Implicit Support Expansion for Offline Reinforcement Learning via Stochastic Policy Optimization","ref_index":9,"is_internal_anchor":false},{"citing_arxiv_id":"2106.01345","citing_title":"Decision Transformer: Reinforcement Learning via Sequence Modeling","ref_index":18,"is_internal_anchor":false},{"citing_arxiv_id":"1910.01708","citing_title":"Benchmarking Batch Deep Reinforcement Learning Algorithms","ref_index":11,"is_internal_anchor":false},{"citing_arxiv_id":"2602.06603","citing_title":"The hidden risks of temporal resampling in clinical reinforcement learning","ref_index":57,"is_internal_anchor":false},{"citing_arxiv_id":"1911.11361","citing_title":"Behavior Regularized Offline Reinforcement Learning","ref_index":11,"is_internal_anchor":false},{"citing_arxiv_id":"2605.11151","citing_title":"RankQ: Offline-to-Online Reinforcement Learning via Self-Supervised Action Ranking","ref_index":10,"is_internal_anchor":false},{"citing_arxiv_id":"2004.07219","citing_title":"D4RL: Datasets for Deep Data-Driven Reinforcement Learning","ref_index":14,"is_internal_anchor":false},{"citing_arxiv_id":"2110.06169","citing_title":"Offline Reinforcement Learning with Implicit Q-Learning","ref_index":8,"is_internal_anchor":false},{"citing_arxiv_id":"1910.00177","citing_title":"Advantage-Weighted Regression: Simple and Scalable Off-Policy Reinforcement Learning","ref_index":9,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/2ISVUUXERZSSO4TPRGY5WHTCDA","json":"https://pith.science/pith/2ISVUUXERZSSO4TPRGY5WHTCDA.json","graph_json":"https://pith.science/api/pith-number/2ISVUUXERZSSO4TPRGY5WHTCDA/graph.json","events_json":"https://pith.science/api/pith-number/2ISVUUXERZSSO4TPRGY5WHTCDA/events.json","paper":"https://pith.science/paper/2ISVUUXE"},"agent_actions":{"view_html":"https://pith.science/pith/2ISVUUXERZSSO4TPRGY5WHTCDA","download_json":"https://pith.science/pith/2ISVUUXERZSSO4TPRGY5WHTCDA.json","view_paper":"https://pith.science/paper/2ISVUUXE","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1906.00949&json=true","fetch_graph":"https://pith.science/api/pith-number/2ISVUUXERZSSO4TPRGY5WHTCDA/graph.json","fetch_events":"https://pith.science/api/pith-number/2ISVUUXERZSSO4TPRGY5WHTCDA/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/2ISVUUXERZSSO4TPRGY5WHTCDA/action/timestamp_anchor","attest_storage":"https://pith.science/pith/2ISVUUXERZSSO4TPRGY5WHTCDA/action/storage_attestation","attest_author":"https://pith.science/pith/2ISVUUXERZSSO4TPRGY5WHTCDA/action/author_attestation","sign_citation":"https://pith.science/pith/2ISVUUXERZSSO4TPRGY5WHTCDA/action/citation_signature","submit_replication":"https://pith.science/pith/2ISVUUXERZSSO4TPRGY5WHTCDA/action/replication_record"}},"created_at":"2026-07-05T00:21:34.797628+00:00","updated_at":"2026-07-05T00:21:34.797628+00:00"}