{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2019:IBNWYLZPA6AZIJ6FWXVDGNGCZR","short_pith_number":"pith:IBNWYLZP","schema_version":"1.0","canonical_sha256":"405b6c2f2f07819427c5b5ea3334c2cc4385ed561695eb1956269f0283ad6920","source":{"kind":"arxiv","id":"1905.06125","version":1},"attestation_state":"computed","paper":{"title":"Distributional Reinforcement Learning for Efficient Exploration","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","stat.ML"],"primary_cat":"cs.LG","authors_text":"Borislav Mavrin, Hengshuai Yao, Kaiwen Wu, Linglong Kong, Shangtong Zhang, Yaoliang Yu","submitted_at":"2019-05-13T19:08:55Z","abstract_excerpt":"In distributional reinforcement learning (RL), the estimated distribution of value function models both the parametric and intrinsic uncertainties. We propose a novel and efficient exploration method for deep RL that has two components. The first is a decaying schedule to suppress the intrinsic uncertainty. The second is an exploration bonus calculated from the upper quantiles of the learned distribution. In Atari 2600 games, our method outperforms QR-DQN in 12 out of 14 hard games (achieving 483 \\% average gain across 49 games in cumulative rewards over QR-DQN with a big win in Venture). We a"},"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":"1905.06125","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2019-05-13T19:08:55Z","cross_cats_sorted":["cs.AI","stat.ML"],"title_canon_sha256":"52e677d89d14e15b66965b453c62c91bf04a4e69bed8b571f3900b0f48b52c8f","abstract_canon_sha256":"c742c0765477aafbf945cf046b6419f0e09800f16143f529fd822799636353f4"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-17T23:46:07.389878Z","signature_b64":"Bd76jYpDmFo9sPz6EUQ19lCgK+HSoUtCPgsi6vOYDp00f+bCohF0CMDRWxErpBlVckQgqdwySc5TxUnSV+uLAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"405b6c2f2f07819427c5b5ea3334c2cc4385ed561695eb1956269f0283ad6920","last_reissued_at":"2026-05-17T23:46:07.389151Z","signature_status":"signed_v1","first_computed_at":"2026-05-17T23:46:07.389151Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Distributional Reinforcement Learning for Efficient Exploration","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","stat.ML"],"primary_cat":"cs.LG","authors_text":"Borislav Mavrin, Hengshuai Yao, Kaiwen Wu, Linglong Kong, Shangtong Zhang, Yaoliang Yu","submitted_at":"2019-05-13T19:08:55Z","abstract_excerpt":"In distributional reinforcement learning (RL), the estimated distribution of value function models both the parametric and intrinsic uncertainties. We propose a novel and efficient exploration method for deep RL that has two components. The first is a decaying schedule to suppress the intrinsic uncertainty. The second is an exploration bonus calculated from the upper quantiles of the learned distribution. In Atari 2600 games, our method outperforms QR-DQN in 12 out of 14 hard games (achieving 483 \\% average gain across 49 games in cumulative rewards over QR-DQN with a big win in Venture). We a"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1905.06125","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":"1905.06125","created_at":"2026-05-17T23:46:07.389248+00:00"},{"alias_kind":"arxiv_version","alias_value":"1905.06125v1","created_at":"2026-05-17T23:46:07.389248+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1905.06125","created_at":"2026-05-17T23:46:07.389248+00:00"},{"alias_kind":"pith_short_12","alias_value":"IBNWYLZPA6AZ","created_at":"2026-05-18T12:33:18.533446+00:00"},{"alias_kind":"pith_short_16","alias_value":"IBNWYLZPA6AZIJ6F","created_at":"2026-05-18T12:33:18.533446+00:00"},{"alias_kind":"pith_short_8","alias_value":"IBNWYLZP","created_at":"2026-05-18T12:33:18.533446+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2409.12917","citing_title":"Training Language Models to Self-Correct via Reinforcement Learning","ref_index":218,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/IBNWYLZPA6AZIJ6FWXVDGNGCZR","json":"https://pith.science/pith/IBNWYLZPA6AZIJ6FWXVDGNGCZR.json","graph_json":"https://pith.science/api/pith-number/IBNWYLZPA6AZIJ6FWXVDGNGCZR/graph.json","events_json":"https://pith.science/api/pith-number/IBNWYLZPA6AZIJ6FWXVDGNGCZR/events.json","paper":"https://pith.science/paper/IBNWYLZP"},"agent_actions":{"view_html":"https://pith.science/pith/IBNWYLZPA6AZIJ6FWXVDGNGCZR","download_json":"https://pith.science/pith/IBNWYLZPA6AZIJ6FWXVDGNGCZR.json","view_paper":"https://pith.science/paper/IBNWYLZP","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1905.06125&json=true","fetch_graph":"https://pith.science/api/pith-number/IBNWYLZPA6AZIJ6FWXVDGNGCZR/graph.json","fetch_events":"https://pith.science/api/pith-number/IBNWYLZPA6AZIJ6FWXVDGNGCZR/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/IBNWYLZPA6AZIJ6FWXVDGNGCZR/action/timestamp_anchor","attest_storage":"https://pith.science/pith/IBNWYLZPA6AZIJ6FWXVDGNGCZR/action/storage_attestation","attest_author":"https://pith.science/pith/IBNWYLZPA6AZIJ6FWXVDGNGCZR/action/author_attestation","sign_citation":"https://pith.science/pith/IBNWYLZPA6AZIJ6FWXVDGNGCZR/action/citation_signature","submit_replication":"https://pith.science/pith/IBNWYLZPA6AZIJ6FWXVDGNGCZR/action/replication_record"}},"created_at":"2026-05-17T23:46:07.389248+00:00","updated_at":"2026-05-17T23:46:07.389248+00:00"}