{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2021:HKP7DS2KIUDVBFXTIDO3PYPWRC","short_pith_number":"pith:HKP7DS2K","canonical_record":{"source":{"id":"2106.06854","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2021-06-12T20:21:38Z","cross_cats_sorted":["cs.AI","stat.ML"],"title_canon_sha256":"6283a0c73e2f174407444f84b4b86d59ba21c7baa267d18c65e4977a7aabe588","abstract_canon_sha256":"f3a76222efe8768124228621b511dcf64b820d521bbd1723a37e6e2f6e8e14b6"},"schema_version":"1.0"},"canonical_sha256":"3a9ff1cb4a45075096f340ddb7e1f688a3d2f2c1a6707c1d4a8920f6698b7a5d","source":{"kind":"arxiv","id":"2106.06854","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2106.06854","created_at":"2026-07-05T07:12:14Z"},{"alias_kind":"arxiv_version","alias_value":"2106.06854v2","created_at":"2026-07-05T07:12:14Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2106.06854","created_at":"2026-07-05T07:12:14Z"},{"alias_kind":"pith_short_12","alias_value":"HKP7DS2KIUDV","created_at":"2026-07-05T07:12:14Z"},{"alias_kind":"pith_short_16","alias_value":"HKP7DS2KIUDVBFXT","created_at":"2026-07-05T07:12:14Z"},{"alias_kind":"pith_short_8","alias_value":"HKP7DS2K","created_at":"2026-07-05T07:12:14Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2021:HKP7DS2KIUDVBFXTIDO3PYPWRC","target":"record","payload":{"canonical_record":{"source":{"id":"2106.06854","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2021-06-12T20:21:38Z","cross_cats_sorted":["cs.AI","stat.ML"],"title_canon_sha256":"6283a0c73e2f174407444f84b4b86d59ba21c7baa267d18c65e4977a7aabe588","abstract_canon_sha256":"f3a76222efe8768124228621b511dcf64b820d521bbd1723a37e6e2f6e8e14b6"},"schema_version":"1.0"},"canonical_sha256":"3a9ff1cb4a45075096f340ddb7e1f688a3d2f2c1a6707c1d4a8920f6698b7a5d","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:12:14.609492Z","signature_b64":"0vTk0yOIiHmui/DSkc6T8CHnk1Dke8AdK7p9XcetfZmgjhNDrMJv6rKjzhJAzr7HlJ6LM0wyzeKYUtnLTuV5Dw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"3a9ff1cb4a45075096f340ddb7e1f688a3d2f2c1a6707c1d4a8920f6698b7a5d","last_reissued_at":"2026-07-05T07:12:14.609029Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:12:14.609029Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2106.06854","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":"2026-07-05T07:12:14Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"zBGED2BJZNSVHzKIgVp+73KNopLBzdacJXF4WeDIyAHwA1yiPpp+v2Wf6izlNy3oUFYG2jc3+AnSzt6mOhQ4AQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-06T08:45:29.583166Z"},"content_sha256":"91c943b76e74ef316a264defc4fccfca487727b8ac386cb49fe12687e044b133","schema_version":"1.0","event_id":"sha256:91c943b76e74ef316a264defc4fccfca487727b8ac386cb49fe12687e044b133"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2021:HKP7DS2KIUDVBFXTIDO3PYPWRC","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"A Deep Reinforcement Learning Approach to Marginalized Importance Sampling with the Successor Representation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","stat.ML"],"primary_cat":"cs.LG","authors_text":"David Meger, Doina Precup, Scott Fujimoto","submitted_at":"2021-06-12T20:21:38Z","abstract_excerpt":"Marginalized importance sampling (MIS), which measures the density ratio between the state-action occupancy of a target policy and that of a sampling distribution, is a promising approach for off-policy evaluation. However, current state-of-the-art MIS methods rely on complex optimization tricks and succeed mostly on simple toy problems. We bridge the gap between MIS and deep reinforcement learning by observing that the density ratio can be computed from the successor representation of the target policy. The successor representation can be trained through deep reinforcement learning methodolog"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2106.06854","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/2106.06854/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"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T07:12:14Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"VY3ol+3CzQu4oGp2SeR5wvrpIEiuyLIPR8CKBaIHY3spfUDC/CuwbaIGgPNd/DaCrFxCR3WE+6lN2vy3PpQzBg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-06T08:45:29.583675Z"},"content_sha256":"d3ea9dc559cf969549f295722c21b82a5fcbeeecf5719ac2300923a0d81ee5ae","schema_version":"1.0","event_id":"sha256:d3ea9dc559cf969549f295722c21b82a5fcbeeecf5719ac2300923a0d81ee5ae"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/HKP7DS2KIUDVBFXTIDO3PYPWRC/bundle.json","state_url":"https://pith.science/pith/HKP7DS2KIUDVBFXTIDO3PYPWRC/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/HKP7DS2KIUDVBFXTIDO3PYPWRC/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":"2026-08-06T08:45:29Z","links":{"resolver":"https://pith.science/pith/HKP7DS2KIUDVBFXTIDO3PYPWRC","bundle":"https://pith.science/pith/HKP7DS2KIUDVBFXTIDO3PYPWRC/bundle.json","state":"https://pith.science/pith/HKP7DS2KIUDVBFXTIDO3PYPWRC/state.json","well_known_bundle":"https://pith.science/.well-known/pith/HKP7DS2KIUDVBFXTIDO3PYPWRC/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2021:HKP7DS2KIUDVBFXTIDO3PYPWRC","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":"f3a76222efe8768124228621b511dcf64b820d521bbd1723a37e6e2f6e8e14b6","cross_cats_sorted":["cs.AI","stat.ML"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2021-06-12T20:21:38Z","title_canon_sha256":"6283a0c73e2f174407444f84b4b86d59ba21c7baa267d18c65e4977a7aabe588"},"schema_version":"1.0","source":{"id":"2106.06854","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2106.06854","created_at":"2026-07-05T07:12:14Z"},{"alias_kind":"arxiv_version","alias_value":"2106.06854v2","created_at":"2026-07-05T07:12:14Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2106.06854","created_at":"2026-07-05T07:12:14Z"},{"alias_kind":"pith_short_12","alias_value":"HKP7DS2KIUDV","created_at":"2026-07-05T07:12:14Z"},{"alias_kind":"pith_short_16","alias_value":"HKP7DS2KIUDVBFXT","created_at":"2026-07-05T07:12:14Z"},{"alias_kind":"pith_short_8","alias_value":"HKP7DS2K","created_at":"2026-07-05T07:12:14Z"}],"graph_snapshots":[{"event_id":"sha256:d3ea9dc559cf969549f295722c21b82a5fcbeeecf5719ac2300923a0d81ee5ae","target":"graph","created_at":"2026-07-05T07:12:14Z","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"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2106.06854/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Marginalized importance sampling (MIS), which measures the density ratio between the state-action occupancy of a target policy and that of a sampling distribution, is a promising approach for off-policy evaluation. However, current state-of-the-art MIS methods rely on complex optimization tricks and succeed mostly on simple toy problems. We bridge the gap between MIS and deep reinforcement learning by observing that the density ratio can be computed from the successor representation of the target policy. The successor representation can be trained through deep reinforcement learning methodolog","authors_text":"David Meger, Doina Precup, Scott Fujimoto","cross_cats":["cs.AI","stat.ML"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2021-06-12T20:21:38Z","title":"A Deep Reinforcement Learning Approach to Marginalized Importance Sampling with the Successor Representation"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2106.06854","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":"sha256:91c943b76e74ef316a264defc4fccfca487727b8ac386cb49fe12687e044b133","target":"record","created_at":"2026-07-05T07:12:14Z","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":"f3a76222efe8768124228621b511dcf64b820d521bbd1723a37e6e2f6e8e14b6","cross_cats_sorted":["cs.AI","stat.ML"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2021-06-12T20:21:38Z","title_canon_sha256":"6283a0c73e2f174407444f84b4b86d59ba21c7baa267d18c65e4977a7aabe588"},"schema_version":"1.0","source":{"id":"2106.06854","kind":"arxiv","version":2}},"canonical_sha256":"3a9ff1cb4a45075096f340ddb7e1f688a3d2f2c1a6707c1d4a8920f6698b7a5d","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"3a9ff1cb4a45075096f340ddb7e1f688a3d2f2c1a6707c1d4a8920f6698b7a5d","first_computed_at":"2026-07-05T07:12:14.609029Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T07:12:14.609029Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"0vTk0yOIiHmui/DSkc6T8CHnk1Dke8AdK7p9XcetfZmgjhNDrMJv6rKjzhJAzr7HlJ6LM0wyzeKYUtnLTuV5Dw==","signature_status":"signed_v1","signed_at":"2026-07-05T07:12:14.609492Z","signed_message":"canonical_sha256_bytes"},"source_id":"2106.06854","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:91c943b76e74ef316a264defc4fccfca487727b8ac386cb49fe12687e044b133","sha256:d3ea9dc559cf969549f295722c21b82a5fcbeeecf5719ac2300923a0d81ee5ae"],"state_sha256":"c1b27c2a67f73c2e0b4042b4837de585b794ae3abf0f0846969e7fa855ceb331"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"PbBWiHd5NDz2W0iABzsnGP6gysucS4/pT7cO3o9aA/VkK90gvwA4DlR9mnZMA66zlXhZOAPatnETy5L/NGcHCA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-06T08:45:29.588550Z","bundle_sha256":"61a3b0f369e94850b5503c5ad115fa11dea661314e74be7326f60a6f87542a52"}}