{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2016:EHIFKRAWID45HVAS6OD7MEJQCS","short_pith_number":"pith:EHIFKRAW","canonical_record":{"source":{"id":"1606.07636","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2016-06-24T10:54:41Z","cross_cats_sorted":["stat.ML"],"title_canon_sha256":"dbb1dabd7ebc87403b96783c6f0f9e16fd4822193e4a4e2a3fc871aae992554b","abstract_canon_sha256":"1355de6b4147b524ec9564f997b60a63bb11890807b6eba2399c6c3e6f562db6"},"schema_version":"1.0"},"canonical_sha256":"21d055441640f9d3d412f387f611301498cdf454eaf52b1191b752942feb8904","source":{"kind":"arxiv","id":"1606.07636","version":3},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1606.07636","created_at":"2026-05-18T00:28:16Z"},{"alias_kind":"arxiv_version","alias_value":"1606.07636v3","created_at":"2026-05-18T00:28:16Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1606.07636","created_at":"2026-05-18T00:28:16Z"},{"alias_kind":"pith_short_12","alias_value":"EHIFKRAWID45","created_at":"2026-05-18T12:30:12Z"},{"alias_kind":"pith_short_16","alias_value":"EHIFKRAWID45HVAS","created_at":"2026-05-18T12:30:12Z"},{"alias_kind":"pith_short_8","alias_value":"EHIFKRAW","created_at":"2026-05-18T12:30:12Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2016:EHIFKRAWID45HVAS6OD7MEJQCS","target":"record","payload":{"canonical_record":{"source":{"id":"1606.07636","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2016-06-24T10:54:41Z","cross_cats_sorted":["stat.ML"],"title_canon_sha256":"dbb1dabd7ebc87403b96783c6f0f9e16fd4822193e4a4e2a3fc871aae992554b","abstract_canon_sha256":"1355de6b4147b524ec9564f997b60a63bb11890807b6eba2399c6c3e6f562db6"},"schema_version":"1.0"},"canonical_sha256":"21d055441640f9d3d412f387f611301498cdf454eaf52b1191b752942feb8904","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-18T00:28:16.613118Z","signature_b64":"P/Pts1De4sSVSYuMemMyJTNmlzX/j4rJVDjkuRyc0Ykm5M6JqyYkyLH8rMX2FgpYB5qXkD27xRBRqlEJORGCAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"21d055441640f9d3d412f387f611301498cdf454eaf52b1191b752942feb8904","last_reissued_at":"2026-05-18T00:28:16.612300Z","signature_status":"signed_v1","first_computed_at":"2026-05-18T00:28:16.612300Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"1606.07636","source_version":3,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-18T00:28:16Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"QVEC/mN5FgAKoEjNT/iHzWZ04h0MgEq4+5cN6Wf6k8R5YGPeWUNoUIXv/bR/6PGnHmHQpoOI5ewFp1avTYPnBA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-01T03:26:37.074738Z"},"content_sha256":"985e6b667b288d90ddc8bd703e2a21f76cce09a252d8518dc230fccc6ef33ead","schema_version":"1.0","event_id":"sha256:985e6b667b288d90ddc8bd703e2a21f76cce09a252d8518dc230fccc6ef33ead"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2016:EHIFKRAWID45HVAS6OD7MEJQCS","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Is the Bellman residual a bad proxy?","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Bilal Piot, Matthieu Geist, Olivier Pietquin","submitted_at":"2016-06-24T10:54:41Z","abstract_excerpt":"This paper aims at theoretically and empirically comparing two standard optimization criteria for Reinforcement Learning: i) maximization of the mean value and ii) minimization of the Bellman residual. For that purpose, we place ourselves in the framework of policy search algorithms, that are usually designed to maximize the mean value, and derive a method that minimizes the residual $\\|T_* v_\\pi - v_\\pi\\|_{1,\\nu}$ over policies. A theoretical analysis shows how good this proxy is to policy optimization, and notably that it is better than its value-based counterpart. We also propose experiment"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1606.07636","kind":"arxiv","version":3},"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"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-18T00:28:16Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Vhc1KXTX+lY9e1idYpSFP2VFVZUl/H+w9RhPh+lHsGJR/xOQgX/0SPl1iwT+d4dyCnBB/hSzS+9jUJueq0A5AQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-01T03:26:37.075092Z"},"content_sha256":"a1a2323c06ce9bf852ec89a3c21b4b0fb30f9acff7158e58673d7b0ba51586c7","schema_version":"1.0","event_id":"sha256:a1a2323c06ce9bf852ec89a3c21b4b0fb30f9acff7158e58673d7b0ba51586c7"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/EHIFKRAWID45HVAS6OD7MEJQCS/bundle.json","state_url":"https://pith.science/pith/EHIFKRAWID45HVAS6OD7MEJQCS/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/EHIFKRAWID45HVAS6OD7MEJQCS/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":"2026-06-01T03:26:37Z","links":{"resolver":"https://pith.science/pith/EHIFKRAWID45HVAS6OD7MEJQCS","bundle":"https://pith.science/pith/EHIFKRAWID45HVAS6OD7MEJQCS/bundle.json","state":"https://pith.science/pith/EHIFKRAWID45HVAS6OD7MEJQCS/state.json","well_known_bundle":"https://pith.science/.well-known/pith/EHIFKRAWID45HVAS6OD7MEJQCS/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2016:EHIFKRAWID45HVAS6OD7MEJQCS","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":"1355de6b4147b524ec9564f997b60a63bb11890807b6eba2399c6c3e6f562db6","cross_cats_sorted":["stat.ML"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2016-06-24T10:54:41Z","title_canon_sha256":"dbb1dabd7ebc87403b96783c6f0f9e16fd4822193e4a4e2a3fc871aae992554b"},"schema_version":"1.0","source":{"id":"1606.07636","kind":"arxiv","version":3}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1606.07636","created_at":"2026-05-18T00:28:16Z"},{"alias_kind":"arxiv_version","alias_value":"1606.07636v3","created_at":"2026-05-18T00:28:16Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1606.07636","created_at":"2026-05-18T00:28:16Z"},{"alias_kind":"pith_short_12","alias_value":"EHIFKRAWID45","created_at":"2026-05-18T12:30:12Z"},{"alias_kind":"pith_short_16","alias_value":"EHIFKRAWID45HVAS","created_at":"2026-05-18T12:30:12Z"},{"alias_kind":"pith_short_8","alias_value":"EHIFKRAW","created_at":"2026-05-18T12:30:12Z"}],"graph_snapshots":[{"event_id":"sha256:a1a2323c06ce9bf852ec89a3c21b4b0fb30f9acff7158e58673d7b0ba51586c7","target":"graph","created_at":"2026-05-18T00:28:16Z","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"},"paper":{"abstract_excerpt":"This paper aims at theoretically and empirically comparing two standard optimization criteria for Reinforcement Learning: i) maximization of the mean value and ii) minimization of the Bellman residual. For that purpose, we place ourselves in the framework of policy search algorithms, that are usually designed to maximize the mean value, and derive a method that minimizes the residual $\\|T_* v_\\pi - v_\\pi\\|_{1,\\nu}$ over policies. A theoretical analysis shows how good this proxy is to policy optimization, and notably that it is better than its value-based counterpart. We also propose experiment","authors_text":"Bilal Piot, Matthieu Geist, Olivier Pietquin","cross_cats":["stat.ML"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2016-06-24T10:54:41Z","title":"Is the Bellman residual a bad proxy?"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1606.07636","kind":"arxiv","version":3},"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":"sha256:985e6b667b288d90ddc8bd703e2a21f76cce09a252d8518dc230fccc6ef33ead","target":"record","created_at":"2026-05-18T00:28:16Z","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":"1355de6b4147b524ec9564f997b60a63bb11890807b6eba2399c6c3e6f562db6","cross_cats_sorted":["stat.ML"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2016-06-24T10:54:41Z","title_canon_sha256":"dbb1dabd7ebc87403b96783c6f0f9e16fd4822193e4a4e2a3fc871aae992554b"},"schema_version":"1.0","source":{"id":"1606.07636","kind":"arxiv","version":3}},"canonical_sha256":"21d055441640f9d3d412f387f611301498cdf454eaf52b1191b752942feb8904","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"21d055441640f9d3d412f387f611301498cdf454eaf52b1191b752942feb8904","first_computed_at":"2026-05-18T00:28:16.612300Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-18T00:28:16.612300Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"P/Pts1De4sSVSYuMemMyJTNmlzX/j4rJVDjkuRyc0Ykm5M6JqyYkyLH8rMX2FgpYB5qXkD27xRBRqlEJORGCAQ==","signature_status":"signed_v1","signed_at":"2026-05-18T00:28:16.613118Z","signed_message":"canonical_sha256_bytes"},"source_id":"1606.07636","source_kind":"arxiv","source_version":3}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:985e6b667b288d90ddc8bd703e2a21f76cce09a252d8518dc230fccc6ef33ead","sha256:a1a2323c06ce9bf852ec89a3c21b4b0fb30f9acff7158e58673d7b0ba51586c7"],"state_sha256":"a373a65e71842859d87f29cfa9ae95e7aa3efb2db53867bfb289c89b68e1c02f"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"8G8LBPrOwyC5NUION3iG5pL2Sbik+T5A494JejMdmOnLRRbg2yF2fzabwzWFZUWBpBOE9s+vb2rZL7q1iht0DA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-06-01T03:26:37.077203Z","bundle_sha256":"6c39e5ee55fb2b934258a219c80049e78462c0935c28761b94e1f2899b9c84c8"}}