{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2019:YFMXA7LKTAXCTZZV2VG2257QBI","short_pith_number":"pith:YFMXA7LK","schema_version":"1.0","canonical_sha256":"c159707d6a982e29e735d54dad77f00a35c0cfba91c2975884ba094ec2b0114a","source":{"kind":"arxiv","id":"1906.08189","version":5},"attestation_state":"computed","paper":{"title":"Reward Prediction Error as an Exploration Objective in Deep RL","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Anthony Bisulco, Ben Eisner, Daniel Lee, Daniel Yang, Eric Mitchell, Riley Simmons-Edler, Sebastian Seung","submitted_at":"2019-06-19T16:06:36Z","abstract_excerpt":"A major challenge in reinforcement learning is exploration, when local dithering methods such as epsilon-greedy sampling are insufficient to solve a given task. Many recent methods have proposed to intrinsically motivate an agent to seek novel states, driving the agent to discover improved reward. However, while state-novelty exploration methods are suitable for tasks where novel observations correlate well with improved reward, they may not explore more efficiently than epsilon-greedy approaches in environments where the two are not well-correlated. In this paper, we distinguish between explo"},"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":"1906.08189","kind":"arxiv","version":5},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2019-06-19T16:06:36Z","cross_cats_sorted":["stat.ML"],"title_canon_sha256":"71ee5dcfa611d804e4f29603e181f31659911d89e4a5723b3aaf9f99f0fcdc40","abstract_canon_sha256":"41171ae6b7f52a745ce4f3e8ae39ee08836c16ee01170826c203fa4f1692dc5d"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T02:06:47.514153Z","signature_b64":"f+RJ6JvV9yik+zSvxAxnO1QLhX6FFUnEQRKQthJ2RFOZrZ3sJENgIQS+kQgyPFXhHf4rM8qWn5sABaa1z4AZCw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"c159707d6a982e29e735d54dad77f00a35c0cfba91c2975884ba094ec2b0114a","last_reissued_at":"2026-07-05T02:06:47.513700Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T02:06:47.513700Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Reward Prediction Error as an Exploration Objective in Deep RL","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Anthony Bisulco, Ben Eisner, Daniel Lee, Daniel Yang, Eric Mitchell, Riley Simmons-Edler, Sebastian Seung","submitted_at":"2019-06-19T16:06:36Z","abstract_excerpt":"A major challenge in reinforcement learning is exploration, when local dithering methods such as epsilon-greedy sampling are insufficient to solve a given task. Many recent methods have proposed to intrinsically motivate an agent to seek novel states, driving the agent to discover improved reward. However, while state-novelty exploration methods are suitable for tasks where novel observations correlate well with improved reward, they may not explore more efficiently than epsilon-greedy approaches in environments where the two are not well-correlated. In this paper, we distinguish between explo"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1906.08189","kind":"arxiv","version":5},"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/1906.08189/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":"1906.08189","created_at":"2026-07-05T02:06:47.513764+00:00"},{"alias_kind":"arxiv_version","alias_value":"1906.08189v5","created_at":"2026-07-05T02:06:47.513764+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1906.08189","created_at":"2026-07-05T02:06:47.513764+00:00"},{"alias_kind":"pith_short_12","alias_value":"YFMXA7LKTAXC","created_at":"2026-07-05T02:06:47.513764+00:00"},{"alias_kind":"pith_short_16","alias_value":"YFMXA7LKTAXCTZZV","created_at":"2026-07-05T02:06:47.513764+00:00"},{"alias_kind":"pith_short_8","alias_value":"YFMXA7LK","created_at":"2026-07-05T02:06:47.513764+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2501.18093","citing_title":"Reward Prediction Error Prioritisation in Experience Replay: The RPE-PER Method","ref_index":22,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/YFMXA7LKTAXCTZZV2VG2257QBI","json":"https://pith.science/pith/YFMXA7LKTAXCTZZV2VG2257QBI.json","graph_json":"https://pith.science/api/pith-number/YFMXA7LKTAXCTZZV2VG2257QBI/graph.json","events_json":"https://pith.science/api/pith-number/YFMXA7LKTAXCTZZV2VG2257QBI/events.json","paper":"https://pith.science/paper/YFMXA7LK"},"agent_actions":{"view_html":"https://pith.science/pith/YFMXA7LKTAXCTZZV2VG2257QBI","download_json":"https://pith.science/pith/YFMXA7LKTAXCTZZV2VG2257QBI.json","view_paper":"https://pith.science/paper/YFMXA7LK","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1906.08189&json=true","fetch_graph":"https://pith.science/api/pith-number/YFMXA7LKTAXCTZZV2VG2257QBI/graph.json","fetch_events":"https://pith.science/api/pith-number/YFMXA7LKTAXCTZZV2VG2257QBI/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/YFMXA7LKTAXCTZZV2VG2257QBI/action/timestamp_anchor","attest_storage":"https://pith.science/pith/YFMXA7LKTAXCTZZV2VG2257QBI/action/storage_attestation","attest_author":"https://pith.science/pith/YFMXA7LKTAXCTZZV2VG2257QBI/action/author_attestation","sign_citation":"https://pith.science/pith/YFMXA7LKTAXCTZZV2VG2257QBI/action/citation_signature","submit_replication":"https://pith.science/pith/YFMXA7LKTAXCTZZV2VG2257QBI/action/replication_record"}},"created_at":"2026-07-05T02:06:47.513764+00:00","updated_at":"2026-07-05T02:06:47.513764+00:00"}