{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2018:LINENCOETPNQYYBQEIJR35ROBN","short_pith_number":"pith:LINENCOE","schema_version":"1.0","canonical_sha256":"5a1a4689c49bdb0c603022131df62e0b573a920678d143cfad5ef5c971493f38","source":{"kind":"arxiv","id":"1807.09647","version":4},"attestation_state":"computed","paper":{"title":"Variational Bayesian Reinforcement Learning with Regret Bounds","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","stat.ML"],"primary_cat":"cs.LG","authors_text":"Brendan O'Donoghue","submitted_at":"2018-07-25T14:56:09Z","abstract_excerpt":"In reinforcement learning the Q-values summarize the expected future rewards that the agent will attain. However, they cannot capture the epistemic uncertainty about those rewards. In this work we derive a new Bellman operator with associated fixed point we call the `knowledge values'. These K-values compress both the expected future rewards and the epistemic uncertainty into a single value, so that high uncertainty, high reward, or both, can yield high K-values. The key principle is to endow the agent with a risk-seeking utility function that is carefully tuned to balance exploration and expl"},"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":"1807.09647","kind":"arxiv","version":4},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2018-07-25T14:56:09Z","cross_cats_sorted":["cs.AI","stat.ML"],"title_canon_sha256":"dc88172380f76a4a2d390d4820c84956722c4ad7965c3726bf26cd65f2a78d8e","abstract_canon_sha256":"19cb7888b2d068c2be135b8a3490d80b4505487dd5222f47dca4e8e3c8f43ce3"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T05:22:40.887682Z","signature_b64":"mJ3PkpgY3jhRS5hXpp5n8mtP44ogEmurajBnF0uFdeXUPdAqow332fdY6lNxIidXQRQ2ieJnAQBf6/fM3zFQBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"5a1a4689c49bdb0c603022131df62e0b573a920678d143cfad5ef5c971493f38","last_reissued_at":"2026-07-05T05:22:40.887222Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T05:22:40.887222Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Variational Bayesian Reinforcement Learning with Regret Bounds","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","stat.ML"],"primary_cat":"cs.LG","authors_text":"Brendan O'Donoghue","submitted_at":"2018-07-25T14:56:09Z","abstract_excerpt":"In reinforcement learning the Q-values summarize the expected future rewards that the agent will attain. However, they cannot capture the epistemic uncertainty about those rewards. In this work we derive a new Bellman operator with associated fixed point we call the `knowledge values'. These K-values compress both the expected future rewards and the epistemic uncertainty into a single value, so that high uncertainty, high reward, or both, can yield high K-values. The key principle is to endow the agent with a risk-seeking utility function that is carefully tuned to balance exploration and expl"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1807.09647","kind":"arxiv","version":4},"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/1807.09647/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":"1807.09647","created_at":"2026-07-05T05:22:40.887278+00:00"},{"alias_kind":"arxiv_version","alias_value":"1807.09647v4","created_at":"2026-07-05T05:22:40.887278+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1807.09647","created_at":"2026-07-05T05:22:40.887278+00:00"},{"alias_kind":"pith_short_12","alias_value":"LINENCOETPNQ","created_at":"2026-07-05T05:22:40.887278+00:00"},{"alias_kind":"pith_short_16","alias_value":"LINENCOETPNQYYBQ","created_at":"2026-07-05T05:22:40.887278+00:00"},{"alias_kind":"pith_short_8","alias_value":"LINENCOE","created_at":"2026-07-05T05:22:40.887278+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/LINENCOETPNQYYBQEIJR35ROBN","json":"https://pith.science/pith/LINENCOETPNQYYBQEIJR35ROBN.json","graph_json":"https://pith.science/api/pith-number/LINENCOETPNQYYBQEIJR35ROBN/graph.json","events_json":"https://pith.science/api/pith-number/LINENCOETPNQYYBQEIJR35ROBN/events.json","paper":"https://pith.science/paper/LINENCOE"},"agent_actions":{"view_html":"https://pith.science/pith/LINENCOETPNQYYBQEIJR35ROBN","download_json":"https://pith.science/pith/LINENCOETPNQYYBQEIJR35ROBN.json","view_paper":"https://pith.science/paper/LINENCOE","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1807.09647&json=true","fetch_graph":"https://pith.science/api/pith-number/LINENCOETPNQYYBQEIJR35ROBN/graph.json","fetch_events":"https://pith.science/api/pith-number/LINENCOETPNQYYBQEIJR35ROBN/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/LINENCOETPNQYYBQEIJR35ROBN/action/timestamp_anchor","attest_storage":"https://pith.science/pith/LINENCOETPNQYYBQEIJR35ROBN/action/storage_attestation","attest_author":"https://pith.science/pith/LINENCOETPNQYYBQEIJR35ROBN/action/author_attestation","sign_citation":"https://pith.science/pith/LINENCOETPNQYYBQEIJR35ROBN/action/citation_signature","submit_replication":"https://pith.science/pith/LINENCOETPNQYYBQEIJR35ROBN/action/replication_record"}},"created_at":"2026-07-05T05:22:40.887278+00:00","updated_at":"2026-07-05T05:22:40.887278+00:00"}