{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:S7CB65RWPISEOTQ2PGUQAM2JXX","short_pith_number":"pith:S7CB65RW","schema_version":"1.0","canonical_sha256":"97c41f76367a24474e1a79a9003349bdd2459195335f38223355f77a2c0a9fba","source":{"kind":"arxiv","id":"2312.04386","version":3},"attestation_state":"computed","paper":{"title":"Model-Based Epistemic Variance of Values for Risk-Aware Policy Optimization","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Alessandro G. Bottero, Carlos E. Luis, Felix Berkenkamp, Jan Peters, Julia Vinogradska","submitted_at":"2023-12-07T15:55:58Z","abstract_excerpt":"We consider the problem of quantifying uncertainty over expected cumulative rewards in model-based reinforcement learning. In particular, we focus on characterizing the variance over values induced by a distribution over Markov decision processes (MDPs). Previous work upper bounds the posterior variance over values by solving a so-called uncertainty Bellman equation (UBE), but the over-approximation may result in inefficient exploration. We propose a new UBE whose solution converges to the true posterior variance over values and leads to lower regret in tabular exploration problems. We identif"},"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":"2312.04386","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2023-12-07T15:55:58Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"bc9e8fb9520c8f35fd1f26a87f89e9f06ce8699dae06b1d44279f44ee75d1654","abstract_canon_sha256":"35e54778d1d9e22faafc8d7da8da5e9fd394f1e05abd35566734cb5ee42560f9"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:07:44.739463Z","signature_b64":"X4MSxE80r8pjRhUvOma4WYCgQ33UC81N/VpjQnUYARpthmaj+XNTQNL+LLQ8VYG43l3bLJdsvNUOYry+/+d1CA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"97c41f76367a24474e1a79a9003349bdd2459195335f38223355f77a2c0a9fba","last_reissued_at":"2026-07-05T09:07:44.738945Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:07:44.738945Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Model-Based Epistemic Variance of Values for Risk-Aware Policy Optimization","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Alessandro G. Bottero, Carlos E. Luis, Felix Berkenkamp, Jan Peters, Julia Vinogradska","submitted_at":"2023-12-07T15:55:58Z","abstract_excerpt":"We consider the problem of quantifying uncertainty over expected cumulative rewards in model-based reinforcement learning. In particular, we focus on characterizing the variance over values induced by a distribution over Markov decision processes (MDPs). Previous work upper bounds the posterior variance over values by solving a so-called uncertainty Bellman equation (UBE), but the over-approximation may result in inefficient exploration. We propose a new UBE whose solution converges to the true posterior variance over values and leads to lower regret in tabular exploration problems. We identif"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2312.04386","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":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2312.04386/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":"2312.04386","created_at":"2026-07-05T09:07:44.739011+00:00"},{"alias_kind":"arxiv_version","alias_value":"2312.04386v3","created_at":"2026-07-05T09:07:44.739011+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2312.04386","created_at":"2026-07-05T09:07:44.739011+00:00"},{"alias_kind":"pith_short_12","alias_value":"S7CB65RWPISE","created_at":"2026-07-05T09:07:44.739011+00:00"},{"alias_kind":"pith_short_16","alias_value":"S7CB65RWPISEOTQ2","created_at":"2026-07-05T09:07:44.739011+00:00"},{"alias_kind":"pith_short_8","alias_value":"S7CB65RW","created_at":"2026-07-05T09:07:44.739011+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2412.20824","citing_title":"Isoperimetry is All We Need: Langevin Posterior Sampling for RL with Sublinear Regret","ref_index":59,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/S7CB65RWPISEOTQ2PGUQAM2JXX","json":"https://pith.science/pith/S7CB65RWPISEOTQ2PGUQAM2JXX.json","graph_json":"https://pith.science/api/pith-number/S7CB65RWPISEOTQ2PGUQAM2JXX/graph.json","events_json":"https://pith.science/api/pith-number/S7CB65RWPISEOTQ2PGUQAM2JXX/events.json","paper":"https://pith.science/paper/S7CB65RW"},"agent_actions":{"view_html":"https://pith.science/pith/S7CB65RWPISEOTQ2PGUQAM2JXX","download_json":"https://pith.science/pith/S7CB65RWPISEOTQ2PGUQAM2JXX.json","view_paper":"https://pith.science/paper/S7CB65RW","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2312.04386&json=true","fetch_graph":"https://pith.science/api/pith-number/S7CB65RWPISEOTQ2PGUQAM2JXX/graph.json","fetch_events":"https://pith.science/api/pith-number/S7CB65RWPISEOTQ2PGUQAM2JXX/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/S7CB65RWPISEOTQ2PGUQAM2JXX/action/timestamp_anchor","attest_storage":"https://pith.science/pith/S7CB65RWPISEOTQ2PGUQAM2JXX/action/storage_attestation","attest_author":"https://pith.science/pith/S7CB65RWPISEOTQ2PGUQAM2JXX/action/author_attestation","sign_citation":"https://pith.science/pith/S7CB65RWPISEOTQ2PGUQAM2JXX/action/citation_signature","submit_replication":"https://pith.science/pith/S7CB65RWPISEOTQ2PGUQAM2JXX/action/replication_record"}},"created_at":"2026-07-05T09:07:44.739011+00:00","updated_at":"2026-07-05T09:07:44.739011+00:00"}