{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2019:BNISRJCMTQP75F7CC2VBNI3TUG","short_pith_number":"pith:BNISRJCM","schema_version":"1.0","canonical_sha256":"0b5128a44c9c1ffe97e216aa16a373a1a90b0ce78995175d792caa3a02b7fef5","source":{"kind":"arxiv","id":"1901.09455","version":1},"attestation_state":"computed","paper":{"title":"Off-Policy Deep Reinforcement Learning by Bootstrapping the Covariate Shift","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Carles Gelada, Marc G. Bellemare","submitted_at":"2019-01-27T22:56:50Z","abstract_excerpt":"In this paper we revisit the method of off-policy corrections for reinforcement learning (COP-TD) pioneered by Hallak et al. (2017). Under this method, online updates to the value function are reweighted to avoid divergence issues typical of off-policy learning. While Hallak et al.'s solution is appealing, it cannot easily be transferred to nonlinear function approximation. First, it requires a projection step onto the probability simplex; second, even though the operator describing the expected behavior of the off-policy learning algorithm is convergent, it is not known to be a contraction ma"},"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":"1901.09455","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2019-01-27T22:56:50Z","cross_cats_sorted":["stat.ML"],"title_canon_sha256":"0b75f90db82ba9f2a58fcbe5ace08bfff93cdda5ef603b11d634301249f61c51","abstract_canon_sha256":"53e97d88bb3a616e50bc1ba6d441bac1c3b1c5c97c049633bd8d464e46612348"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-17T23:55:24.696388Z","signature_b64":"JfiFpi7wRdfwo7IH51fbw18lRJTVtp1Fu/CdW1BUGbeZ7I8o48WxcTraursEvgFGqr1NgyAAWRTV5YQkVAcMDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"0b5128a44c9c1ffe97e216aa16a373a1a90b0ce78995175d792caa3a02b7fef5","last_reissued_at":"2026-05-17T23:55:24.695881Z","signature_status":"signed_v1","first_computed_at":"2026-05-17T23:55:24.695881Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Off-Policy Deep Reinforcement Learning by Bootstrapping the Covariate Shift","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Carles Gelada, Marc G. Bellemare","submitted_at":"2019-01-27T22:56:50Z","abstract_excerpt":"In this paper we revisit the method of off-policy corrections for reinforcement learning (COP-TD) pioneered by Hallak et al. (2017). Under this method, online updates to the value function are reweighted to avoid divergence issues typical of off-policy learning. While Hallak et al.'s solution is appealing, it cannot easily be transferred to nonlinear function approximation. First, it requires a projection step onto the probability simplex; second, even though the operator describing the expected behavior of the off-policy learning algorithm is convergent, it is not known to be a contraction ma"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1901.09455","kind":"arxiv","version":1},"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"},"aliases":[{"alias_kind":"arxiv","alias_value":"1901.09455","created_at":"2026-05-17T23:55:24.695980+00:00"},{"alias_kind":"arxiv_version","alias_value":"1901.09455v1","created_at":"2026-05-17T23:55:24.695980+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1901.09455","created_at":"2026-05-17T23:55:24.695980+00:00"},{"alias_kind":"pith_short_12","alias_value":"BNISRJCMTQP7","created_at":"2026-05-18T12:33:12.712433+00:00"},{"alias_kind":"pith_short_16","alias_value":"BNISRJCMTQP75F7C","created_at":"2026-05-18T12:33:12.712433+00:00"},{"alias_kind":"pith_short_8","alias_value":"BNISRJCM","created_at":"2026-05-18T12:33:12.712433+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"1907.00456","citing_title":"Way Off-Policy Batch Deep Reinforcement Learning of Implicit Human Preferences in Dialog","ref_index":19,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/BNISRJCMTQP75F7CC2VBNI3TUG","json":"https://pith.science/pith/BNISRJCMTQP75F7CC2VBNI3TUG.json","graph_json":"https://pith.science/api/pith-number/BNISRJCMTQP75F7CC2VBNI3TUG/graph.json","events_json":"https://pith.science/api/pith-number/BNISRJCMTQP75F7CC2VBNI3TUG/events.json","paper":"https://pith.science/paper/BNISRJCM"},"agent_actions":{"view_html":"https://pith.science/pith/BNISRJCMTQP75F7CC2VBNI3TUG","download_json":"https://pith.science/pith/BNISRJCMTQP75F7CC2VBNI3TUG.json","view_paper":"https://pith.science/paper/BNISRJCM","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1901.09455&json=true","fetch_graph":"https://pith.science/api/pith-number/BNISRJCMTQP75F7CC2VBNI3TUG/graph.json","fetch_events":"https://pith.science/api/pith-number/BNISRJCMTQP75F7CC2VBNI3TUG/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/BNISRJCMTQP75F7CC2VBNI3TUG/action/timestamp_anchor","attest_storage":"https://pith.science/pith/BNISRJCMTQP75F7CC2VBNI3TUG/action/storage_attestation","attest_author":"https://pith.science/pith/BNISRJCMTQP75F7CC2VBNI3TUG/action/author_attestation","sign_citation":"https://pith.science/pith/BNISRJCMTQP75F7CC2VBNI3TUG/action/citation_signature","submit_replication":"https://pith.science/pith/BNISRJCMTQP75F7CC2VBNI3TUG/action/replication_record"}},"created_at":"2026-05-17T23:55:24.695980+00:00","updated_at":"2026-05-17T23:55:24.695980+00:00"}