{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:D74UIN55W4FCUQM2Q5AOWYOBM5","short_pith_number":"pith:D74UIN55","canonical_record":{"source":{"id":"2505.11478","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2025-05-16T17:40:01Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"3d7fdb6aed13e19f27af651e79bb7a1bb89553f213d404c6d0fe5b649e2ef33f","abstract_canon_sha256":"70b53be951adb6677d0bec43d882455c0d5da62d4fd33e84916bbff87c440fa6"},"schema_version":"1.0"},"canonical_sha256":"1ff94437bdb70a2a419a8740eb61c16742c9f8e0c0aba016e36c28dca6f78e26","source":{"kind":"arxiv","id":"2505.11478","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2505.11478","created_at":"2026-07-05T12:07:38Z"},{"alias_kind":"arxiv_version","alias_value":"2505.11478v2","created_at":"2026-07-05T12:07:38Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.11478","created_at":"2026-07-05T12:07:38Z"},{"alias_kind":"pith_short_12","alias_value":"D74UIN55W4FC","created_at":"2026-07-05T12:07:38Z"},{"alias_kind":"pith_short_16","alias_value":"D74UIN55W4FCUQM2","created_at":"2026-07-05T12:07:38Z"},{"alias_kind":"pith_short_8","alias_value":"D74UIN55","created_at":"2026-07-05T12:07:38Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:D74UIN55W4FCUQM2Q5AOWYOBM5","target":"record","payload":{"canonical_record":{"source":{"id":"2505.11478","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2025-05-16T17:40:01Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"3d7fdb6aed13e19f27af651e79bb7a1bb89553f213d404c6d0fe5b649e2ef33f","abstract_canon_sha256":"70b53be951adb6677d0bec43d882455c0d5da62d4fd33e84916bbff87c440fa6"},"schema_version":"1.0"},"canonical_sha256":"1ff94437bdb70a2a419a8740eb61c16742c9f8e0c0aba016e36c28dca6f78e26","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T12:07:38.419107Z","signature_b64":"UbwsZGOTPjEr85YbQoa6c9zPTd/jBhVl1y9ROdq/tFGVmBJfSrzDXCVuiJCfUctrVBfEIDytOhIBgZdFF/77CQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"1ff94437bdb70a2a419a8740eb61c16742c9f8e0c0aba016e36c28dca6f78e26","last_reissued_at":"2026-07-05T12:07:38.418417Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T12:07:38.418417Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2505.11478","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":"2026-07-05T12:07:38Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"tT3yuNPgYTPmw8rr98V/UL4rniFhQfq+WzxG8evcj1htghXG+s7LOFNlWZpSEByrMQg6dDvS9/41w+G3nuKIAQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-08T22:06:55.744895Z"},"content_sha256":"69c191fd6588cf436173419ec425caeebb0df2e6d03cae5f5d468d694f0e0654","schema_version":"1.0","event_id":"sha256:69c191fd6588cf436173419ec425caeebb0df2e6d03cae5f5d468d694f0e0654"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:D74UIN55W4FCUQM2Q5AOWYOBM5","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Automatic Reward Shaping from Confounded Offline Data","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.AI","authors_text":"Elias Bareinboim, Junzhe Zhang, Mingxuan Li","submitted_at":"2025-05-16T17:40:01Z","abstract_excerpt":"A key task in Artificial Intelligence is learning effective policies for controlling agents in unknown environments to optimize performance measures. Off-policy learning methods, like Q-learning, allow learners to make optimal decisions based on past experiences. This paper studies off-policy learning from biased data in complex and high-dimensional domains where \\emph{unobserved confounding} cannot be ruled out a priori. Building on the well-celebrated Deep Q-Network (DQN), we propose a novel deep reinforcement learning algorithm robust to confounding biases in observed data. Specifically, ou"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.11478","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/2505.11478/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":"2026-07-05T12:07:38Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"TjGWe0Sm/V9obDlLbiQqzUn3LHR4Xa2bsMTKELcixqQ/WLCIHKctV3T2rUsKnQDfY8rYhwilNqorAccO+vHsBw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-08T22:06:55.745419Z"},"content_sha256":"b72dde990ca653d7796f75ec2930b340439a30c2c927bc492ec1b40292246fa7","schema_version":"1.0","event_id":"sha256:b72dde990ca653d7796f75ec2930b340439a30c2c927bc492ec1b40292246fa7"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/D74UIN55W4FCUQM2Q5AOWYOBM5/bundle.json","state_url":"https://pith.science/pith/D74UIN55W4FCUQM2Q5AOWYOBM5/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/D74UIN55W4FCUQM2Q5AOWYOBM5/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":"2026-08-08T22:06:55Z","links":{"resolver":"https://pith.science/pith/D74UIN55W4FCUQM2Q5AOWYOBM5","bundle":"https://pith.science/pith/D74UIN55W4FCUQM2Q5AOWYOBM5/bundle.json","state":"https://pith.science/pith/D74UIN55W4FCUQM2Q5AOWYOBM5/state.json","well_known_bundle":"https://pith.science/.well-known/pith/D74UIN55W4FCUQM2Q5AOWYOBM5/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:D74UIN55W4FCUQM2Q5AOWYOBM5","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":"70b53be951adb6677d0bec43d882455c0d5da62d4fd33e84916bbff87c440fa6","cross_cats_sorted":["cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2025-05-16T17:40:01Z","title_canon_sha256":"3d7fdb6aed13e19f27af651e79bb7a1bb89553f213d404c6d0fe5b649e2ef33f"},"schema_version":"1.0","source":{"id":"2505.11478","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2505.11478","created_at":"2026-07-05T12:07:38Z"},{"alias_kind":"arxiv_version","alias_value":"2505.11478v2","created_at":"2026-07-05T12:07:38Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.11478","created_at":"2026-07-05T12:07:38Z"},{"alias_kind":"pith_short_12","alias_value":"D74UIN55W4FC","created_at":"2026-07-05T12:07:38Z"},{"alias_kind":"pith_short_16","alias_value":"D74UIN55W4FCUQM2","created_at":"2026-07-05T12:07:38Z"},{"alias_kind":"pith_short_8","alias_value":"D74UIN55","created_at":"2026-07-05T12:07:38Z"}],"graph_snapshots":[{"event_id":"sha256:b72dde990ca653d7796f75ec2930b340439a30c2c927bc492ec1b40292246fa7","target":"graph","created_at":"2026-07-05T12:07:38Z","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/2505.11478/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"A key task in Artificial Intelligence is learning effective policies for controlling agents in unknown environments to optimize performance measures. Off-policy learning methods, like Q-learning, allow learners to make optimal decisions based on past experiences. This paper studies off-policy learning from biased data in complex and high-dimensional domains where \\emph{unobserved confounding} cannot be ruled out a priori. Building on the well-celebrated Deep Q-Network (DQN), we propose a novel deep reinforcement learning algorithm robust to confounding biases in observed data. Specifically, ou","authors_text":"Elias Bareinboim, Junzhe Zhang, Mingxuan Li","cross_cats":["cs.LG"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2025-05-16T17:40:01Z","title":"Automatic Reward Shaping from Confounded Offline Data"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.11478","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":"sha256:69c191fd6588cf436173419ec425caeebb0df2e6d03cae5f5d468d694f0e0654","target":"record","created_at":"2026-07-05T12:07:38Z","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":"70b53be951adb6677d0bec43d882455c0d5da62d4fd33e84916bbff87c440fa6","cross_cats_sorted":["cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2025-05-16T17:40:01Z","title_canon_sha256":"3d7fdb6aed13e19f27af651e79bb7a1bb89553f213d404c6d0fe5b649e2ef33f"},"schema_version":"1.0","source":{"id":"2505.11478","kind":"arxiv","version":2}},"canonical_sha256":"1ff94437bdb70a2a419a8740eb61c16742c9f8e0c0aba016e36c28dca6f78e26","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"1ff94437bdb70a2a419a8740eb61c16742c9f8e0c0aba016e36c28dca6f78e26","first_computed_at":"2026-07-05T12:07:38.418417Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T12:07:38.418417Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"UbwsZGOTPjEr85YbQoa6c9zPTd/jBhVl1y9ROdq/tFGVmBJfSrzDXCVuiJCfUctrVBfEIDytOhIBgZdFF/77CQ==","signature_status":"signed_v1","signed_at":"2026-07-05T12:07:38.419107Z","signed_message":"canonical_sha256_bytes"},"source_id":"2505.11478","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:69c191fd6588cf436173419ec425caeebb0df2e6d03cae5f5d468d694f0e0654","sha256:b72dde990ca653d7796f75ec2930b340439a30c2c927bc492ec1b40292246fa7"],"state_sha256":"6290db007929857472ed0df83ae3390eb87b4d377c19d3d5bec6242676414955"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"1G8v1W/tUt6ibVQifAGovPqbEGZw+NI0ajLZgjXd6WQN2pp/6iyHGYG/o6VLtlMRjYAg/tOwLOn4CjJHxI82Cg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-08T22:06:55.749519Z","bundle_sha256":"24420e0da8f6776f0d6f69752a88f91a2cebe2c12ebc1744ae46d6b7068ee1dc"}}