{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2019:TTCAXWKNNHEEERLWX32TQW4MZN","short_pith_number":"pith:TTCAXWKN","canonical_record":{"source":{"id":"1910.01708","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2019-10-03T20:15:55Z","cross_cats_sorted":["cs.AI","stat.ML"],"title_canon_sha256":"b288aedd284dd5187b25b9614e70486df3cd2fa1a6c57e096f443186d963459a","abstract_canon_sha256":"52bc3739e1831baf67e8a14a5a40e7555d91c254c91c18d67b8f2c5ea670ac68"},"schema_version":"1.0"},"canonical_sha256":"9cc40bd94d69c8424576bef5385b8ccb5ae43f3e8fcd2a64912a0a6985a1e466","source":{"kind":"arxiv","id":"1910.01708","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1910.01708","created_at":"2026-05-17T23:38:13Z"},{"alias_kind":"arxiv_version","alias_value":"1910.01708v1","created_at":"2026-05-17T23:38:13Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1910.01708","created_at":"2026-05-17T23:38:13Z"},{"alias_kind":"pith_short_12","alias_value":"TTCAXWKNNHEE","created_at":"2026-05-18T12:33:30Z"},{"alias_kind":"pith_short_16","alias_value":"TTCAXWKNNHEEERLW","created_at":"2026-05-18T12:33:30Z"},{"alias_kind":"pith_short_8","alias_value":"TTCAXWKN","created_at":"2026-05-18T12:33:30Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2019:TTCAXWKNNHEEERLWX32TQW4MZN","target":"record","payload":{"canonical_record":{"source":{"id":"1910.01708","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2019-10-03T20:15:55Z","cross_cats_sorted":["cs.AI","stat.ML"],"title_canon_sha256":"b288aedd284dd5187b25b9614e70486df3cd2fa1a6c57e096f443186d963459a","abstract_canon_sha256":"52bc3739e1831baf67e8a14a5a40e7555d91c254c91c18d67b8f2c5ea670ac68"},"schema_version":"1.0"},"canonical_sha256":"9cc40bd94d69c8424576bef5385b8ccb5ae43f3e8fcd2a64912a0a6985a1e466","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-17T23:38:13.297517Z","signature_b64":"lEePs2yBoSR9FJ4qqy3S5AfOM9B9VOmvZOvZdgkyAvaPmc15Mh3nZEGBo5mg0omrXlMTHJHhuTIO1U3XBblPDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"9cc40bd94d69c8424576bef5385b8ccb5ae43f3e8fcd2a64912a0a6985a1e466","last_reissued_at":"2026-05-17T23:38:13.296922Z","signature_status":"signed_v1","first_computed_at":"2026-05-17T23:38:13.296922Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"1910.01708","source_version":1,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-17T23:38:13Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Fbuj60N/NHaU2S4PlYgZU3c+wxHQxwViNShqd3Efj+k/C+s+4EADEyxwKULgpyCAGAu9b/xLqpv94MQH0TOjDw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-19T00:08:37.010530Z"},"content_sha256":"95869cd2b4fa36e2bbae8d79fd12edabe05601f48d255781030ddf8f00f8ab67","schema_version":"1.0","event_id":"sha256:95869cd2b4fa36e2bbae8d79fd12edabe05601f48d255781030ddf8f00f8ab67"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2019:TTCAXWKNNHEEERLWX32TQW4MZN","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Benchmarking Batch Deep Reinforcement Learning Algorithms","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"Many batch deep RL algorithms underperform online DQN and the behavioral policy itself when trained on fixed Atari data from one policy.","cross_cats":["cs.AI","stat.ML"],"primary_cat":"cs.LG","authors_text":"Edoardo Conti, Joelle Pineau, Mohammad Ghavamzadeh, Scott Fujimoto","submitted_at":"2019-10-03T20:15:55Z","abstract_excerpt":"Widely-used deep reinforcement learning algorithms have been shown to fail in the batch setting--learning from a fixed data set without interaction with the environment. Following this result, there have been several papers showing reasonable performances under a variety of environments and batch settings. In this paper, we benchmark the performance of recent off-policy and batch reinforcement learning algorithms under unified settings on the Atari domain, with data generated by a single partially-trained behavioral policy. We find that under these conditions, many of these algorithms underper"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"under these conditions, many of these algorithms underperform DQN trained online with the same amount of data, as well as the partially-trained behavioral policy. ... we adapt the Batch-Constrained Q-learning algorithm to a discrete-action setting, and show it outperforms all existing algorithms at this task.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"That data generated by a single partially-trained behavioral policy under unified settings produces a representative and fair testbed for comparing batch RL algorithms.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"Many batch RL algorithms underperform both online DQN and the behavioral policy on Atari; an adapted discrete-action BCQ outperforms the others tested.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"Many batch deep RL algorithms underperform online DQN and the behavioral policy itself when trained on fixed Atari data from one policy.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"a166490578a6cf086f90d485b3ccaede0b58f6148046db76d51db23d88f1f8ad"},"source":{"id":"1910.01708","kind":"arxiv","version":1},"verdict":{"id":"b0a6fd49-69b5-4df4-b81d-9d426bf70cc4","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-17T19:24:29.450533Z","strongest_claim":"under these conditions, many of these algorithms underperform DQN trained online with the same amount of data, as well as the partially-trained behavioral policy. ... we adapt the Batch-Constrained Q-learning algorithm to a discrete-action setting, and show it outperforms all existing algorithms at this task.","one_line_summary":"Many batch RL algorithms underperform both online DQN and the behavioral policy on Atari; an adapted discrete-action BCQ outperforms the others tested.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"That data generated by a single partially-trained behavioral policy under unified settings produces a representative and fair testbed for comparing batch RL algorithms.","pith_extraction_headline":"Many batch deep RL algorithms underperform online DQN and the behavioral policy itself when trained on fixed Atari data from one policy."},"references":{"count":18,"sample":[{"doi":"","year":1907,"title":"Striving for simplicity in off-policy deep reinforcement learning.arXiv preprint arXiv:1907.04543","work_id":"588a4dc4-2821-4960-ab47-4a85390287c3","ref_index":1,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":null,"title":"Exploration by random network distillation","work_id":"5a87fef6-96e2-4d5b-91ec-1a7c9a43cab9","ref_index":2,"cited_arxiv_id":"1810.12894","is_internal_anchor":true},{"doi":"","year":null,"title":"Dopamine: A research framework for deep reinforcement learning","work_id":"a0c1e29f-e775-49dc-977e-bbbca0daa6d8","ref_index":3,"cited_arxiv_id":"1812.06110","is_internal_anchor":true},{"doi":"","year":null,"title":"Bellemare, and R´ emi Munos","work_id":"f3531420-1352-4be7-b9b7-229fb79f889d","ref_index":4,"cited_arxiv_id":"1710.10044","is_internal_anchor":true},{"doi":"","year":2052,"title":"Off-policy deep reinforcement learning without exploration","work_id":"9bd09ac4-5e91-4d50-9e9b-38436443b511","ref_index":5,"cited_arxiv_id":"","is_internal_anchor":false}],"resolved_work":18,"snapshot_sha256":"cf15819122f07fb9170bf0b70ce830d22d3a93022edc2c398111ffc6da057e4c","internal_anchors":7},"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"},"verdict_id":"b0a6fd49-69b5-4df4-b81d-9d426bf70cc4"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-17T23:38:13Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"oyiMtgFvpTM9at7Vvf0b+PwVVlbDKi3VuBvFQEkJp8jK8WpzISdQVRfFLtHRKgoYbKDFzXju0psTvTJxRhCpDQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-19T00:08:37.011731Z"},"content_sha256":"efc52b9a2cc94d347fd0dcb959f5f87c3e233c1f275589bdf87112a5daf2fccd","schema_version":"1.0","event_id":"sha256:efc52b9a2cc94d347fd0dcb959f5f87c3e233c1f275589bdf87112a5daf2fccd"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/TTCAXWKNNHEEERLWX32TQW4MZN/bundle.json","state_url":"https://pith.science/pith/TTCAXWKNNHEEERLWX32TQW4MZN/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/TTCAXWKNNHEEERLWX32TQW4MZN/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-05-19T00:08:37Z","links":{"resolver":"https://pith.science/pith/TTCAXWKNNHEEERLWX32TQW4MZN","bundle":"https://pith.science/pith/TTCAXWKNNHEEERLWX32TQW4MZN/bundle.json","state":"https://pith.science/pith/TTCAXWKNNHEEERLWX32TQW4MZN/state.json","well_known_bundle":"https://pith.science/.well-known/pith/TTCAXWKNNHEEERLWX32TQW4MZN/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2019:TTCAXWKNNHEEERLWX32TQW4MZN","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"52bc3739e1831baf67e8a14a5a40e7555d91c254c91c18d67b8f2c5ea670ac68","cross_cats_sorted":["cs.AI","stat.ML"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2019-10-03T20:15:55Z","title_canon_sha256":"b288aedd284dd5187b25b9614e70486df3cd2fa1a6c57e096f443186d963459a"},"schema_version":"1.0","source":{"id":"1910.01708","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1910.01708","created_at":"2026-05-17T23:38:13Z"},{"alias_kind":"arxiv_version","alias_value":"1910.01708v1","created_at":"2026-05-17T23:38:13Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1910.01708","created_at":"2026-05-17T23:38:13Z"},{"alias_kind":"pith_short_12","alias_value":"TTCAXWKNNHEE","created_at":"2026-05-18T12:33:30Z"},{"alias_kind":"pith_short_16","alias_value":"TTCAXWKNNHEEERLW","created_at":"2026-05-18T12:33:30Z"},{"alias_kind":"pith_short_8","alias_value":"TTCAXWKN","created_at":"2026-05-18T12:33:30Z"}],"graph_snapshots":[{"event_id":"sha256:efc52b9a2cc94d347fd0dcb959f5f87c3e233c1f275589bdf87112a5daf2fccd","target":"graph","created_at":"2026-05-17T23:38:13Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":4,"items":[{"attestation":"unclaimed","claim_id":"C1","kind":"strongest_claim","source":"verdict.strongest_claim","status":"machine_extracted","text":"under these conditions, many of these algorithms underperform DQN trained online with the same amount of data, as well as the partially-trained behavioral policy. ... we adapt the Batch-Constrained Q-learning algorithm to a discrete-action setting, and show it outperforms all existing algorithms at this task."},{"attestation":"unclaimed","claim_id":"C2","kind":"weakest_assumption","source":"verdict.weakest_assumption","status":"machine_extracted","text":"That data generated by a single partially-trained behavioral policy under unified settings produces a representative and fair testbed for comparing batch RL algorithms."},{"attestation":"unclaimed","claim_id":"C3","kind":"one_line_summary","source":"verdict.one_line_summary","status":"machine_extracted","text":"Many batch RL algorithms underperform both online DQN and the behavioral policy on Atari; an adapted discrete-action BCQ outperforms the others tested."},{"attestation":"unclaimed","claim_id":"C4","kind":"headline","source":"verdict.pith_extraction.headline","status":"machine_extracted","text":"Many batch deep RL algorithms underperform online DQN and the behavioral policy itself when trained on fixed Atari data from one policy."}],"snapshot_sha256":"a166490578a6cf086f90d485b3ccaede0b58f6148046db76d51db23d88f1f8ad"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"paper":{"abstract_excerpt":"Widely-used deep reinforcement learning algorithms have been shown to fail in the batch setting--learning from a fixed data set without interaction with the environment. Following this result, there have been several papers showing reasonable performances under a variety of environments and batch settings. In this paper, we benchmark the performance of recent off-policy and batch reinforcement learning algorithms under unified settings on the Atari domain, with data generated by a single partially-trained behavioral policy. We find that under these conditions, many of these algorithms underper","authors_text":"Edoardo Conti, Joelle Pineau, Mohammad Ghavamzadeh, Scott Fujimoto","cross_cats":["cs.AI","stat.ML"],"headline":"Many batch deep RL algorithms underperform online DQN and the behavioral policy itself when trained on fixed Atari data from one policy.","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2019-10-03T20:15:55Z","title":"Benchmarking Batch Deep Reinforcement Learning Algorithms"},"references":{"count":18,"internal_anchors":7,"resolved_work":18,"sample":[{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":1,"title":"Striving for simplicity in off-policy deep reinforcement learning.arXiv preprint arXiv:1907.04543","work_id":"588a4dc4-2821-4960-ab47-4a85390287c3","year":1907},{"cited_arxiv_id":"1810.12894","doi":"","is_internal_anchor":true,"ref_index":2,"title":"Exploration by random network distillation","work_id":"5a87fef6-96e2-4d5b-91ec-1a7c9a43cab9","year":null},{"cited_arxiv_id":"1812.06110","doi":"","is_internal_anchor":true,"ref_index":3,"title":"Dopamine: A research framework for deep reinforcement learning","work_id":"a0c1e29f-e775-49dc-977e-bbbca0daa6d8","year":null},{"cited_arxiv_id":"1710.10044","doi":"","is_internal_anchor":true,"ref_index":4,"title":"Bellemare, and R´ emi Munos","work_id":"f3531420-1352-4be7-b9b7-229fb79f889d","year":null},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":5,"title":"Off-policy deep reinforcement learning without exploration","work_id":"9bd09ac4-5e91-4d50-9e9b-38436443b511","year":2052}],"snapshot_sha256":"cf15819122f07fb9170bf0b70ce830d22d3a93022edc2c398111ffc6da057e4c"},"source":{"id":"1910.01708","kind":"arxiv","version":1},"verdict":{"created_at":"2026-05-17T19:24:29.450533Z","id":"b0a6fd49-69b5-4df4-b81d-9d426bf70cc4","model_set":{"reader":"grok-4.3"},"one_line_summary":"Many batch RL algorithms underperform both online DQN and the behavioral policy on Atari; an adapted discrete-action BCQ outperforms the others tested.","pipeline_version":"pith-pipeline@v0.9.0","pith_extraction_headline":"Many batch deep RL algorithms underperform online DQN and the behavioral policy itself when trained on fixed Atari data from one policy.","strongest_claim":"under these conditions, many of these algorithms underperform DQN trained online with the same amount of data, as well as the partially-trained behavioral policy. ... we adapt the Batch-Constrained Q-learning algorithm to a discrete-action setting, and show it outperforms all existing algorithms at this task.","weakest_assumption":"That data generated by a single partially-trained behavioral policy under unified settings produces a representative and fair testbed for comparing batch RL algorithms."}},"verdict_id":"b0a6fd49-69b5-4df4-b81d-9d426bf70cc4"}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:95869cd2b4fa36e2bbae8d79fd12edabe05601f48d255781030ddf8f00f8ab67","target":"record","created_at":"2026-05-17T23:38:13Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"52bc3739e1831baf67e8a14a5a40e7555d91c254c91c18d67b8f2c5ea670ac68","cross_cats_sorted":["cs.AI","stat.ML"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2019-10-03T20:15:55Z","title_canon_sha256":"b288aedd284dd5187b25b9614e70486df3cd2fa1a6c57e096f443186d963459a"},"schema_version":"1.0","source":{"id":"1910.01708","kind":"arxiv","version":1}},"canonical_sha256":"9cc40bd94d69c8424576bef5385b8ccb5ae43f3e8fcd2a64912a0a6985a1e466","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"9cc40bd94d69c8424576bef5385b8ccb5ae43f3e8fcd2a64912a0a6985a1e466","first_computed_at":"2026-05-17T23:38:13.296922Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-17T23:38:13.296922Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"lEePs2yBoSR9FJ4qqy3S5AfOM9B9VOmvZOvZdgkyAvaPmc15Mh3nZEGBo5mg0omrXlMTHJHhuTIO1U3XBblPDA==","signature_status":"signed_v1","signed_at":"2026-05-17T23:38:13.297517Z","signed_message":"canonical_sha256_bytes"},"source_id":"1910.01708","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:95869cd2b4fa36e2bbae8d79fd12edabe05601f48d255781030ddf8f00f8ab67","sha256:efc52b9a2cc94d347fd0dcb959f5f87c3e233c1f275589bdf87112a5daf2fccd"],"state_sha256":"38c246d34d1719471d0de0ff1237e5a3b924f5896f7b5a6dcc9d43c31fcc6f95"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"jCmug6udb100Y+ZIQpL02dTqgBgYtV1hqStO0EYwIzRXVkZNSZepgx3xZxJTE6fdm+GQkKl4DOyzdCgAwyUSCg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-05-19T00:08:37.014917Z","bundle_sha256":"3c7876b9aacb341b97c335dd46123725f26350d8917778fda0c47bd703515851"}}