{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:HQSDMYTU2HN5LZUNPAVLVW72YV","short_pith_number":"pith:HQSDMYTU","schema_version":"1.0","canonical_sha256":"3c24366274d1dbd5e68d782abadbfac574968cd3066aa3270acd6a6a01a71343","source":{"kind":"arxiv","id":"2305.20081","version":2},"attestation_state":"computed","paper":{"title":"Efficient Diffusion Policies for Offline Reinforcement Learning","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Bingyi Kang, Chao Du, Shuicheng Yan, Tianyu Pang, Xiao Ma","submitted_at":"2023-05-31T17:55:21Z","abstract_excerpt":"Offline reinforcement learning (RL) aims to learn optimal policies from offline datasets, where the parameterization of policies is crucial but often overlooked. Recently, Diffsuion-QL significantly boosts the performance of offline RL by representing a policy with a diffusion model, whose success relies on a parametrized Markov Chain with hundreds of steps for sampling. However, Diffusion-QL suffers from two critical limitations. 1) It is computationally inefficient to forward and backward through the whole Markov chain during training. 2) It is incompatible with maximum likelihood-based RL a"},"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":"2305.20081","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.LG","submitted_at":"2023-05-31T17:55:21Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"817776701ac7f6852343c99861c55921f70b742b8adc85321148dfd06ef7d207","abstract_canon_sha256":"2c5c761972742fb38ed3e76936409a99f9bf8355787bb37a42db31989fc65dcf"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:05:31.213291Z","signature_b64":"jlB90M3vtIFhnlrOrDsqIl/yCr0j7rNgH/C00NRrfMa/UG7QYhpnVs91Co4i66J7+RF7BDH8DXlA/Dh4HL0tAg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"3c24366274d1dbd5e68d782abadbfac574968cd3066aa3270acd6a6a01a71343","last_reissued_at":"2026-07-05T07:05:31.212849Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:05:31.212849Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Efficient Diffusion Policies for Offline Reinforcement Learning","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Bingyi Kang, Chao Du, Shuicheng Yan, Tianyu Pang, Xiao Ma","submitted_at":"2023-05-31T17:55:21Z","abstract_excerpt":"Offline reinforcement learning (RL) aims to learn optimal policies from offline datasets, where the parameterization of policies is crucial but often overlooked. Recently, Diffsuion-QL significantly boosts the performance of offline RL by representing a policy with a diffusion model, whose success relies on a parametrized Markov Chain with hundreds of steps for sampling. However, Diffusion-QL suffers from two critical limitations. 1) It is computationally inefficient to forward and backward through the whole Markov chain during training. 2) It is incompatible with maximum likelihood-based RL a"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2305.20081","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/2305.20081/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":"2305.20081","created_at":"2026-07-05T07:05:31.212905+00:00"},{"alias_kind":"arxiv_version","alias_value":"2305.20081v2","created_at":"2026-07-05T07:05:31.212905+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2305.20081","created_at":"2026-07-05T07:05:31.212905+00:00"},{"alias_kind":"pith_short_12","alias_value":"HQSDMYTU2HN5","created_at":"2026-07-05T07:05:31.212905+00:00"},{"alias_kind":"pith_short_16","alias_value":"HQSDMYTU2HN5LZUN","created_at":"2026-07-05T07:05:31.212905+00:00"},{"alias_kind":"pith_short_8","alias_value":"HQSDMYTU","created_at":"2026-07-05T07:05:31.212905+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":3,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.10825","citing_title":"MODIP: Efficient Model-Based Optimization for Diffusion Policies","ref_index":9,"is_internal_anchor":false},{"citing_arxiv_id":"2606.01151","citing_title":"Lagrangian Perturbation Diffusion Steering: Latent Reinforcement Learning for Generative Policies","ref_index":13,"is_internal_anchor":false},{"citing_arxiv_id":"2506.20882","citing_title":"Resilience Through Escalation: A Graph-Based PACE Architecture for Satellite Threat Response","ref_index":18,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/HQSDMYTU2HN5LZUNPAVLVW72YV","json":"https://pith.science/pith/HQSDMYTU2HN5LZUNPAVLVW72YV.json","graph_json":"https://pith.science/api/pith-number/HQSDMYTU2HN5LZUNPAVLVW72YV/graph.json","events_json":"https://pith.science/api/pith-number/HQSDMYTU2HN5LZUNPAVLVW72YV/events.json","paper":"https://pith.science/paper/HQSDMYTU"},"agent_actions":{"view_html":"https://pith.science/pith/HQSDMYTU2HN5LZUNPAVLVW72YV","download_json":"https://pith.science/pith/HQSDMYTU2HN5LZUNPAVLVW72YV.json","view_paper":"https://pith.science/paper/HQSDMYTU","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2305.20081&json=true","fetch_graph":"https://pith.science/api/pith-number/HQSDMYTU2HN5LZUNPAVLVW72YV/graph.json","fetch_events":"https://pith.science/api/pith-number/HQSDMYTU2HN5LZUNPAVLVW72YV/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/HQSDMYTU2HN5LZUNPAVLVW72YV/action/timestamp_anchor","attest_storage":"https://pith.science/pith/HQSDMYTU2HN5LZUNPAVLVW72YV/action/storage_attestation","attest_author":"https://pith.science/pith/HQSDMYTU2HN5LZUNPAVLVW72YV/action/author_attestation","sign_citation":"https://pith.science/pith/HQSDMYTU2HN5LZUNPAVLVW72YV/action/citation_signature","submit_replication":"https://pith.science/pith/HQSDMYTU2HN5LZUNPAVLVW72YV/action/replication_record"}},"created_at":"2026-07-05T07:05:31.212905+00:00","updated_at":"2026-07-05T07:05:31.212905+00:00"}