{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2016:RJE2KYQOVMYRYVJBGFWBAXQQQT","short_pith_number":"pith:RJE2KYQO","canonical_record":{"source":{"id":"1606.02647","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2016-06-08T17:34:13Z","cross_cats_sorted":["cs.AI","stat.ML"],"title_canon_sha256":"53dcd0f04cc0fdcee2027f8049052538f36225d75892542d3cfecfd1ac62138c","abstract_canon_sha256":"edafccd8972d592c7e581c3d25c66d6a8ca4de8947143cbb2003ee771b384c59"},"schema_version":"1.0"},"canonical_sha256":"8a49a5620eab311c5521316c105e1084d34287d5bd917f2f7bac03ea7911b933","source":{"kind":"arxiv","id":"1606.02647","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1606.02647","created_at":"2026-05-18T00:59:58Z"},{"alias_kind":"arxiv_version","alias_value":"1606.02647v2","created_at":"2026-05-18T00:59:58Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1606.02647","created_at":"2026-05-18T00:59:58Z"},{"alias_kind":"pith_short_12","alias_value":"RJE2KYQOVMYR","created_at":"2026-05-18T12:30:41Z"},{"alias_kind":"pith_short_16","alias_value":"RJE2KYQOVMYRYVJB","created_at":"2026-05-18T12:30:41Z"},{"alias_kind":"pith_short_8","alias_value":"RJE2KYQO","created_at":"2026-05-18T12:30:41Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2016:RJE2KYQOVMYRYVJBGFWBAXQQQT","target":"record","payload":{"canonical_record":{"source":{"id":"1606.02647","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2016-06-08T17:34:13Z","cross_cats_sorted":["cs.AI","stat.ML"],"title_canon_sha256":"53dcd0f04cc0fdcee2027f8049052538f36225d75892542d3cfecfd1ac62138c","abstract_canon_sha256":"edafccd8972d592c7e581c3d25c66d6a8ca4de8947143cbb2003ee771b384c59"},"schema_version":"1.0"},"canonical_sha256":"8a49a5620eab311c5521316c105e1084d34287d5bd917f2f7bac03ea7911b933","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-18T00:59:58.244132Z","signature_b64":"pXjo5qwngQTgltfZ0TaSSU7R9xFarhVL6R7zhXq5GWXv3xIQ+0nYj/jMytGChvrTb81FwKa9ZafkWJmERduOBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"8a49a5620eab311c5521316c105e1084d34287d5bd917f2f7bac03ea7911b933","last_reissued_at":"2026-05-18T00:59:58.243497Z","signature_status":"signed_v1","first_computed_at":"2026-05-18T00:59:58.243497Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"1606.02647","source_version":2,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-18T00:59:58Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"wofjI8pZVHJyX5BCB+uceEbeBkiIAvypNL6tUBecsNwARLVK/7H1U3ub42agHbUNA8USbokGaDWqe28mZbafCw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-30T19:40:04.041858Z"},"content_sha256":"e4236b60af98cec436e9e23a8dfd12bd72ecdbb46bb3b340ff1370bf4e01f947","schema_version":"1.0","event_id":"sha256:e4236b60af98cec436e9e23a8dfd12bd72ecdbb46bb3b340ff1370bf4e01f947"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2016:RJE2KYQOVMYRYVJBGFWBAXQQQT","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Safe and Efficient Off-Policy Reinforcement Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","stat.ML"],"primary_cat":"cs.LG","authors_text":"Anna Harutyunyan, Marc G. Bellemare, R\\'emi Munos, Tom Stepleton","submitted_at":"2016-06-08T17:34:13Z","abstract_excerpt":"In this work, we take a fresh look at some old and new algorithms for off-policy, return-based reinforcement learning. Expressing these in a common form, we derive a novel algorithm, Retrace($\\lambda$), with three desired properties: (1) it has low variance; (2) it safely uses samples collected from any behaviour policy, whatever its degree of \"off-policyness\"; and (3) it is efficient as it makes the best use of samples collected from near on-policy behaviour policies. We analyze the contractive nature of the related operator under both off-policy policy evaluation and control settings and der"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1606.02647","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":""},"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"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-18T00:59:58Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"qZKgwl+9jAXrGAy7sYpiJp3UHQYmGt6nD3xVbtxr4SkOF2Fe5JnMrFu9n41cwJ1Vs3zetDjWQVzbzPleIuNNDQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-30T19:40:04.042528Z"},"content_sha256":"61554cf9626b14ed411f5b66f207741289202b05c12f430eb479623c46544eca","schema_version":"1.0","event_id":"sha256:61554cf9626b14ed411f5b66f207741289202b05c12f430eb479623c46544eca"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/RJE2KYQOVMYRYVJBGFWBAXQQQT/bundle.json","state_url":"https://pith.science/pith/RJE2KYQOVMYRYVJBGFWBAXQQQT/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/RJE2KYQOVMYRYVJBGFWBAXQQQT/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":"2026-05-30T19:40:04Z","links":{"resolver":"https://pith.science/pith/RJE2KYQOVMYRYVJBGFWBAXQQQT","bundle":"https://pith.science/pith/RJE2KYQOVMYRYVJBGFWBAXQQQT/bundle.json","state":"https://pith.science/pith/RJE2KYQOVMYRYVJBGFWBAXQQQT/state.json","well_known_bundle":"https://pith.science/.well-known/pith/RJE2KYQOVMYRYVJBGFWBAXQQQT/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2016:RJE2KYQOVMYRYVJBGFWBAXQQQT","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":"edafccd8972d592c7e581c3d25c66d6a8ca4de8947143cbb2003ee771b384c59","cross_cats_sorted":["cs.AI","stat.ML"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2016-06-08T17:34:13Z","title_canon_sha256":"53dcd0f04cc0fdcee2027f8049052538f36225d75892542d3cfecfd1ac62138c"},"schema_version":"1.0","source":{"id":"1606.02647","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1606.02647","created_at":"2026-05-18T00:59:58Z"},{"alias_kind":"arxiv_version","alias_value":"1606.02647v2","created_at":"2026-05-18T00:59:58Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1606.02647","created_at":"2026-05-18T00:59:58Z"},{"alias_kind":"pith_short_12","alias_value":"RJE2KYQOVMYR","created_at":"2026-05-18T12:30:41Z"},{"alias_kind":"pith_short_16","alias_value":"RJE2KYQOVMYRYVJB","created_at":"2026-05-18T12:30:41Z"},{"alias_kind":"pith_short_8","alias_value":"RJE2KYQO","created_at":"2026-05-18T12:30:41Z"}],"graph_snapshots":[{"event_id":"sha256:61554cf9626b14ed411f5b66f207741289202b05c12f430eb479623c46544eca","target":"graph","created_at":"2026-05-18T00:59:58Z","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":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"paper":{"abstract_excerpt":"In this work, we take a fresh look at some old and new algorithms for off-policy, return-based reinforcement learning. Expressing these in a common form, we derive a novel algorithm, Retrace($\\lambda$), with three desired properties: (1) it has low variance; (2) it safely uses samples collected from any behaviour policy, whatever its degree of \"off-policyness\"; and (3) it is efficient as it makes the best use of samples collected from near on-policy behaviour policies. We analyze the contractive nature of the related operator under both off-policy policy evaluation and control settings and der","authors_text":"Anna Harutyunyan, Marc G. Bellemare, R\\'emi Munos, Tom Stepleton","cross_cats":["cs.AI","stat.ML"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2016-06-08T17:34:13Z","title":"Safe and Efficient Off-Policy Reinforcement Learning"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1606.02647","kind":"arxiv","version":2},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:e4236b60af98cec436e9e23a8dfd12bd72ecdbb46bb3b340ff1370bf4e01f947","target":"record","created_at":"2026-05-18T00:59:58Z","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":"edafccd8972d592c7e581c3d25c66d6a8ca4de8947143cbb2003ee771b384c59","cross_cats_sorted":["cs.AI","stat.ML"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2016-06-08T17:34:13Z","title_canon_sha256":"53dcd0f04cc0fdcee2027f8049052538f36225d75892542d3cfecfd1ac62138c"},"schema_version":"1.0","source":{"id":"1606.02647","kind":"arxiv","version":2}},"canonical_sha256":"8a49a5620eab311c5521316c105e1084d34287d5bd917f2f7bac03ea7911b933","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"8a49a5620eab311c5521316c105e1084d34287d5bd917f2f7bac03ea7911b933","first_computed_at":"2026-05-18T00:59:58.243497Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-18T00:59:58.243497Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"pXjo5qwngQTgltfZ0TaSSU7R9xFarhVL6R7zhXq5GWXv3xIQ+0nYj/jMytGChvrTb81FwKa9ZafkWJmERduOBg==","signature_status":"signed_v1","signed_at":"2026-05-18T00:59:58.244132Z","signed_message":"canonical_sha256_bytes"},"source_id":"1606.02647","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:e4236b60af98cec436e9e23a8dfd12bd72ecdbb46bb3b340ff1370bf4e01f947","sha256:61554cf9626b14ed411f5b66f207741289202b05c12f430eb479623c46544eca"],"state_sha256":"3ed71568ed8c06f5ab5fc7f4622642301a624364b8691325103d6b699cc1449e"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"6bUTxeMydqe18p1ZG+pOsrP8gMWTpDQ6b1iENIJAY4w65m/qcQIEg5CwnFeF0ItqGQJKFKo0TQhcxhCRb3uaBg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-05-30T19:40:04.045887Z","bundle_sha256":"dc4aca58e1e574d9593c9ec5fc226cbe5204c1216c12cec39dbc160337a93bac"}}