{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2020:BS4VGKKUL26RWZ45KEDPBWV2OH","short_pith_number":"pith:BS4VGKKU","schema_version":"1.0","canonical_sha256":"0cb95329545ebd1b679d5106f0daba71d993df9d3be8192e13b75e38223571ca","source":{"kind":"arxiv","id":"2006.04779","version":3},"attestation_state":"computed","paper":{"title":"Conservative Q-Learning for Offline Reinforcement Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Aurick Zhou, Aviral Kumar, George Tucker, Sergey Levine","submitted_at":"2020-06-08T17:53:42Z","abstract_excerpt":"Effectively leveraging large, previously collected datasets in reinforcement learning (RL) is a key challenge for large-scale real-world applications. Offline RL algorithms promise to learn effective policies from previously-collected, static datasets without further interaction. However, in practice, offline RL presents a major challenge, and standard off-policy RL methods can fail due to overestimation of values induced by the distributional shift between the dataset and the learned policy, especially when training on complex and multi-modal data distributions. In this paper, we propose cons"},"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":"2006.04779","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2020-06-08T17:53:42Z","cross_cats_sorted":["stat.ML"],"title_canon_sha256":"ea62a204edfb6c5daa349cac7ed817799f81a519e60dd375eca40a62228e3149","abstract_canon_sha256":"ebef52ba6e8c969fe889154f4f0d4e8c09ed1642ac1e58d2a587f1db12b473fc"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T01:28:20.357910Z","signature_b64":"18BvxUfMq5BRDLigqyf0bHs/XLIs/oG+qCyj1jUogfp2LiBACymKj1Uo6+hTTxwDKtB4NHhdDxJWLJpQa55RBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"0cb95329545ebd1b679d5106f0daba71d993df9d3be8192e13b75e38223571ca","last_reissued_at":"2026-07-05T01:28:20.357403Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T01:28:20.357403Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Conservative Q-Learning for Offline Reinforcement Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Aurick Zhou, Aviral Kumar, George Tucker, Sergey Levine","submitted_at":"2020-06-08T17:53:42Z","abstract_excerpt":"Effectively leveraging large, previously collected datasets in reinforcement learning (RL) is a key challenge for large-scale real-world applications. Offline RL algorithms promise to learn effective policies from previously-collected, static datasets without further interaction. However, in practice, offline RL presents a major challenge, and standard off-policy RL methods can fail due to overestimation of values induced by the distributional shift between the dataset and the learned policy, especially when training on complex and multi-modal data distributions. In this paper, we propose cons"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2006.04779","kind":"arxiv","version":3},"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/2006.04779/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":"2006.04779","created_at":"2026-07-05T01:28:20.357464+00:00"},{"alias_kind":"arxiv_version","alias_value":"2006.04779v3","created_at":"2026-07-05T01:28:20.357464+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2006.04779","created_at":"2026-07-05T01:28:20.357464+00:00"},{"alias_kind":"pith_short_12","alias_value":"BS4VGKKUL26R","created_at":"2026-07-05T01:28:20.357464+00:00"},{"alias_kind":"pith_short_16","alias_value":"BS4VGKKUL26RWZ45","created_at":"2026-07-05T01:28:20.357464+00:00"},{"alias_kind":"pith_short_8","alias_value":"BS4VGKKU","created_at":"2026-07-05T01:28:20.357464+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":19,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.21271","citing_title":"Reward-free Pretraining for Reinforcement Learning via Occupancy Coverage Maximization","ref_index":48,"is_internal_anchor":false},{"citing_arxiv_id":"2607.02092","citing_title":"Guided Action Flow: Q-Guided Inference for Flow-Matching Vision-Language-Action Policies","ref_index":43,"is_internal_anchor":false},{"citing_arxiv_id":"2606.27475","citing_title":"Support-Constrained RL Enables Real-World Policy Improvement without Real-World Experience","ref_index":31,"is_internal_anchor":false},{"citing_arxiv_id":"2606.27865","citing_title":"From Bootstrapping to Sequence Modeling: A Unified Generative Framework for Personalized Landing-Page Modeling","ref_index":20,"is_internal_anchor":false},{"citing_arxiv_id":"2606.30627","citing_title":"Pessimism's Paradox: Conservative Offline Training Amplifies Reward Hacking During Online Adaptation in Reasoning Models","ref_index":11,"is_internal_anchor":false},{"citing_arxiv_id":"2605.22711","citing_title":"Abstraction for Offline Goal-Conditioned Reinforcement Learning","ref_index":25,"is_internal_anchor":false},{"citing_arxiv_id":"2605.11151","citing_title":"RankQ: Offline-to-Online Reinforcement Learning via Self-Supervised Action Ranking","ref_index":3,"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":12,"is_internal_anchor":false},{"citing_arxiv_id":"2605.16826","citing_title":"Decoupling KL and Trajectories: A Unified Perspective for SFT, DAgger, Offline RL, and OPD in LLM Distillation","ref_index":20,"is_internal_anchor":false},{"citing_arxiv_id":"2509.19538","citing_title":"DAWM: Diffusion Action World Models for Offline Reinforcement Learning via Action-Inferred Transitions","ref_index":18,"is_internal_anchor":false},{"citing_arxiv_id":"2512.03847","citing_title":"DVPO: Distributional Value Modeling-based Policy Optimization for LLM Post-Training","ref_index":13,"is_internal_anchor":false},{"citing_arxiv_id":"2602.06603","citing_title":"The hidden risks of temporal resampling in clinical reinforcement learning","ref_index":62,"is_internal_anchor":false},{"citing_arxiv_id":"2603.15759","citing_title":"Simulation Distillation: Pretraining World Models in Simulation for Rapid Real-World Adaptation","ref_index":28,"is_internal_anchor":false},{"citing_arxiv_id":"2108.03298","citing_title":"What Matters in Learning from Offline Human Demonstrations for Robot Manipulation","ref_index":27,"is_internal_anchor":false},{"citing_arxiv_id":"2605.11151","citing_title":"RankQ: Offline-to-Online Reinforcement Learning via Self-Supervised Action Ranking","ref_index":3,"is_internal_anchor":false},{"citing_arxiv_id":"2110.06169","citing_title":"Offline Reinforcement Learning with Implicit Q-Learning","ref_index":9,"is_internal_anchor":false},{"citing_arxiv_id":"2604.05845","citing_title":"JD-BP: A Joint-Decision Generative Framework for Auto-Bidding and Pricing","ref_index":22,"is_internal_anchor":false},{"citing_arxiv_id":"2605.01567","citing_title":"Feedback-Normalized Developer Memory for Reinforcement-Learning Coding Agents: A Safety-Gated MCP Architecture","ref_index":16,"is_internal_anchor":false},{"citing_arxiv_id":"2605.02112","citing_title":"An adaptive variance estimator for relative sparsity","ref_index":79,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/BS4VGKKUL26RWZ45KEDPBWV2OH","json":"https://pith.science/pith/BS4VGKKUL26RWZ45KEDPBWV2OH.json","graph_json":"https://pith.science/api/pith-number/BS4VGKKUL26RWZ45KEDPBWV2OH/graph.json","events_json":"https://pith.science/api/pith-number/BS4VGKKUL26RWZ45KEDPBWV2OH/events.json","paper":"https://pith.science/paper/BS4VGKKU"},"agent_actions":{"view_html":"https://pith.science/pith/BS4VGKKUL26RWZ45KEDPBWV2OH","download_json":"https://pith.science/pith/BS4VGKKUL26RWZ45KEDPBWV2OH.json","view_paper":"https://pith.science/paper/BS4VGKKU","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2006.04779&json=true","fetch_graph":"https://pith.science/api/pith-number/BS4VGKKUL26RWZ45KEDPBWV2OH/graph.json","fetch_events":"https://pith.science/api/pith-number/BS4VGKKUL26RWZ45KEDPBWV2OH/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/BS4VGKKUL26RWZ45KEDPBWV2OH/action/timestamp_anchor","attest_storage":"https://pith.science/pith/BS4VGKKUL26RWZ45KEDPBWV2OH/action/storage_attestation","attest_author":"https://pith.science/pith/BS4VGKKUL26RWZ45KEDPBWV2OH/action/author_attestation","sign_citation":"https://pith.science/pith/BS4VGKKUL26RWZ45KEDPBWV2OH/action/citation_signature","submit_replication":"https://pith.science/pith/BS4VGKKUL26RWZ45KEDPBWV2OH/action/replication_record"}},"created_at":"2026-07-05T01:28:20.357464+00:00","updated_at":"2026-07-05T01:28:20.357464+00:00"}