{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2021:AIGVCECJDA3SHJGRMXCCYAXP6R","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":"bf28fecda263fab97a72713a3827c2fa3c4f7a9ee41b11add606a3c1a174c248","cross_cats_sorted":["stat.ML"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2021-05-21T06:22:34Z","title_canon_sha256":"5a6e30ac6ca390ba2b6c5bb1c21fec3ce545148082a55aa40418f8d8c83d6b74"},"schema_version":"1.0","source":{"id":"2105.10148","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2105.10148","created_at":"2026-07-05T05:20:58Z"},{"alias_kind":"arxiv_version","alias_value":"2105.10148v2","created_at":"2026-07-05T05:20:58Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2105.10148","created_at":"2026-07-05T05:20:58Z"},{"alias_kind":"pith_short_12","alias_value":"AIGVCECJDA3S","created_at":"2026-07-05T05:20:58Z"},{"alias_kind":"pith_short_16","alias_value":"AIGVCECJDA3SHJGR","created_at":"2026-07-05T05:20:58Z"},{"alias_kind":"pith_short_8","alias_value":"AIGVCECJ","created_at":"2026-07-05T05:20:58Z"}],"graph_snapshots":[{"event_id":"sha256:30d94977d9c5767ba941fc3ba4aa137f79597088f3cd37af20d3d87da9056f13","target":"graph","created_at":"2026-07-05T05:20:58Z","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"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2105.10148/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"We show that the popular reinforcement learning (RL) strategy of estimating the state-action value (Q-function) by minimizing the mean squared Bellman error leads to a regression problem with confounding, the inputs and output noise being correlated. Hence, direct minimization of the Bellman error can result in significantly biased Q-function estimates. We explain why fixing the target Q-network in Deep Q-Networks and Fitted Q Evaluation provides a way of overcoming this confounding, thus shedding new light on this popular but not well understood trick in the deep RL literature. An alternative","authors_text":"Arnaud Doucet, Arthur Gretton, Caglar Gulcehre, Liyuan Xu, Nando de Freitas, Tom Le Paine, Yutian Chen","cross_cats":["stat.ML"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2021-05-21T06:22:34Z","title":"On Instrumental Variable Regression for Deep Offline Policy Evaluation"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2105.10148","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":"sha256:9af468b7005ba5e93c18646f4f5ced6f4d48736f5d4deec5e9acc2caf67dce00","target":"record","created_at":"2026-07-05T05:20:58Z","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":"bf28fecda263fab97a72713a3827c2fa3c4f7a9ee41b11add606a3c1a174c248","cross_cats_sorted":["stat.ML"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2021-05-21T06:22:34Z","title_canon_sha256":"5a6e30ac6ca390ba2b6c5bb1c21fec3ce545148082a55aa40418f8d8c83d6b74"},"schema_version":"1.0","source":{"id":"2105.10148","kind":"arxiv","version":2}},"canonical_sha256":"020d511049183723a4d165c42c02eff47a48959f930b8bfa97fba6d7ddeff2ee","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"020d511049183723a4d165c42c02eff47a48959f930b8bfa97fba6d7ddeff2ee","first_computed_at":"2026-07-05T05:20:58.270065Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T05:20:58.270065Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"HVgrV41Bt8id0Jfb2KNKxHlFvVmi2CK0bZ081ec8FUIwjJe2fJyByTPv7vR1DS3z99aaZ9cqY0UyMo4a8SyQAw==","signature_status":"signed_v1","signed_at":"2026-07-05T05:20:58.271234Z","signed_message":"canonical_sha256_bytes"},"source_id":"2105.10148","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:9af468b7005ba5e93c18646f4f5ced6f4d48736f5d4deec5e9acc2caf67dce00","sha256:30d94977d9c5767ba941fc3ba4aa137f79597088f3cd37af20d3d87da9056f13"],"state_sha256":"08328d615d3f3ee3352d1f689347406bb3d3f1f227007aa6035b8e81d2d9216f"}