{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2020:O5XEB4P7S6NBSPPOLRW6NGGJGR","short_pith_number":"pith:O5XEB4P7","schema_version":"1.0","canonical_sha256":"776e40f1ff979a193dee5c6de698c9347a5c800f81dae44ec9bc9484a195e3bd","source":{"kind":"arxiv","id":"2011.01075","version":2},"attestation_state":"computed","paper":{"title":"A Variant of the Wang-Foster-Kakade Lower Bound for the Discounted Setting","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","stat.ML"],"primary_cat":"cs.LG","authors_text":"Nan Jiang, Philip Amortila, Tengyang Xie","submitted_at":"2020-11-02T16:04:42Z","abstract_excerpt":"Recently, Wang et al. (2020) showed a highly intriguing hardness result for batch reinforcement learning (RL) with linearly realizable value function and good feature coverage in the finite-horizon case. In this note we show that once adapted to the discounted setting, the construction can be simplified to a 2-state MDP with 1-dimensional features, such that learning is impossible even with an infinite amount of data."},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2011.01075","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2020-11-02T16:04:42Z","cross_cats_sorted":["cs.AI","stat.ML"],"title_canon_sha256":"ab1e2b23aca469b0f6dbfec1c1263900d7fdecfa998ff54a0c88ce05a073fe4b","abstract_canon_sha256":"fb4f78cbbf3251ed1c6337f235b394197239f8d1232a9924a5d036897e5ae044"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T01:49:06.203689Z","signature_b64":"ZhlLlNoGoxdqIG+HhwvSqQkeJQJBDruOuvltSg0x5yjwlsoOmKbNDSJnmhst1epqkYZs7drlyRhEUzDfwIzGCA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"776e40f1ff979a193dee5c6de698c9347a5c800f81dae44ec9bc9484a195e3bd","last_reissued_at":"2026-07-05T01:49:06.203270Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T01:49:06.203270Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"A Variant of the Wang-Foster-Kakade Lower Bound for the Discounted Setting","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","stat.ML"],"primary_cat":"cs.LG","authors_text":"Nan Jiang, Philip Amortila, Tengyang Xie","submitted_at":"2020-11-02T16:04:42Z","abstract_excerpt":"Recently, Wang et al. (2020) showed a highly intriguing hardness result for batch reinforcement learning (RL) with linearly realizable value function and good feature coverage in the finite-horizon case. In this note we show that once adapted to the discounted setting, the construction can be simplified to a 2-state MDP with 1-dimensional features, such that learning is impossible even with an infinite amount of data."},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2011.01075","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/2011.01075/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"},"aliases":[{"alias_kind":"arxiv","alias_value":"2011.01075","created_at":"2026-07-05T01:49:06.203339+00:00"},{"alias_kind":"arxiv_version","alias_value":"2011.01075v2","created_at":"2026-07-05T01:49:06.203339+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2011.01075","created_at":"2026-07-05T01:49:06.203339+00:00"},{"alias_kind":"pith_short_12","alias_value":"O5XEB4P7S6NB","created_at":"2026-07-05T01:49:06.203339+00:00"},{"alias_kind":"pith_short_16","alias_value":"O5XEB4P7S6NBSPPO","created_at":"2026-07-05T01:49:06.203339+00:00"},{"alias_kind":"pith_short_8","alias_value":"O5XEB4P7","created_at":"2026-07-05T01:49:06.203339+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":5,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2607.05375","citing_title":"Fitted Occupancy-Ratio Evaluation without Bellman Completeness","ref_index":91,"is_internal_anchor":true},{"citing_arxiv_id":"2509.21172","citing_title":"Inverse Reinforcement Learning with Just Classification and a Few Regressions","ref_index":3,"is_internal_anchor":false},{"citing_arxiv_id":"2512.23927","citing_title":"Stationary Reweighting Yields Local Convergence of Soft Fitted Q-Iteration","ref_index":1,"is_internal_anchor":false},{"citing_arxiv_id":"2512.23694","citing_title":"Bellman Calibration for $V$-Learning in Offline Reinforcement Learning","ref_index":1,"is_internal_anchor":false},{"citing_arxiv_id":"2512.23805","citing_title":"Fitted $Q$ Evaluation Without Bellman Completeness via Stationary Weighting","ref_index":1,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/O5XEB4P7S6NBSPPOLRW6NGGJGR","json":"https://pith.science/pith/O5XEB4P7S6NBSPPOLRW6NGGJGR.json","graph_json":"https://pith.science/api/pith-number/O5XEB4P7S6NBSPPOLRW6NGGJGR/graph.json","events_json":"https://pith.science/api/pith-number/O5XEB4P7S6NBSPPOLRW6NGGJGR/events.json","paper":"https://pith.science/paper/O5XEB4P7"},"agent_actions":{"view_html":"https://pith.science/pith/O5XEB4P7S6NBSPPOLRW6NGGJGR","download_json":"https://pith.science/pith/O5XEB4P7S6NBSPPOLRW6NGGJGR.json","view_paper":"https://pith.science/paper/O5XEB4P7","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2011.01075&json=true","fetch_graph":"https://pith.science/api/pith-number/O5XEB4P7S6NBSPPOLRW6NGGJGR/graph.json","fetch_events":"https://pith.science/api/pith-number/O5XEB4P7S6NBSPPOLRW6NGGJGR/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/O5XEB4P7S6NBSPPOLRW6NGGJGR/action/timestamp_anchor","attest_storage":"https://pith.science/pith/O5XEB4P7S6NBSPPOLRW6NGGJGR/action/storage_attestation","attest_author":"https://pith.science/pith/O5XEB4P7S6NBSPPOLRW6NGGJGR/action/author_attestation","sign_citation":"https://pith.science/pith/O5XEB4P7S6NBSPPOLRW6NGGJGR/action/citation_signature","submit_replication":"https://pith.science/pith/O5XEB4P7S6NBSPPOLRW6NGGJGR/action/replication_record"}},"created_at":"2026-07-05T01:49:06.203339+00:00","updated_at":"2026-07-05T01:49:06.203339+00:00"}