{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:WYLLQ42RUU6MX2WMBQPAD6HYBM","short_pith_number":"pith:WYLLQ42R","schema_version":"1.0","canonical_sha256":"b616b87351a53ccbeacc0c1e01f8f80b3597f79492c02b7a62cb2e23abb2b772","source":{"kind":"arxiv","id":"2301.12579","version":2},"attestation_state":"computed","paper":{"title":"Sample Efficient Deep Reinforcement Learning via Local Planning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Botao Hao, Csaba Szepesvari, Dong Yin, Nevena Lazic, Nived Rajaraman, Sridhar Thiagarajan","submitted_at":"2023-01-29T23:17:26Z","abstract_excerpt":"The focus of this work is sample-efficient deep reinforcement learning (RL) with a simulator. One useful property of simulators is that it is typically easy to reset the environment to a previously observed state. We propose an algorithmic framework, named uncertainty-first local planning (UFLP), that takes advantage of this property. Concretely, in each data collection iteration, with some probability, our meta-algorithm resets the environment to an observed state which has high uncertainty, instead of sampling according to the initial-state distribution. The agent-environment interaction the"},"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":"2301.12579","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2023-01-29T23:17:26Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"9ea7b8a6ba08dcd2666be68ab9c4d94fe63b1cff28bc2986123f6c277ef90099","abstract_canon_sha256":"8ae42db02e7d4cb9e35344b010fc4c7873ab8f6945f1db1d5f89c32f0dfab988"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T06:26:45.379550Z","signature_b64":"LD5a1E0SoNk4CZ2GMnW8snh9OIny5MG3c0s4XQSuXqmZLX5E+iVZOWqZUK49yCFiouc8CHDInJ2qQw14xhofAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"b616b87351a53ccbeacc0c1e01f8f80b3597f79492c02b7a62cb2e23abb2b772","last_reissued_at":"2026-07-05T06:26:45.378982Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T06:26:45.378982Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Sample Efficient Deep Reinforcement Learning via Local Planning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Botao Hao, Csaba Szepesvari, Dong Yin, Nevena Lazic, Nived Rajaraman, Sridhar Thiagarajan","submitted_at":"2023-01-29T23:17:26Z","abstract_excerpt":"The focus of this work is sample-efficient deep reinforcement learning (RL) with a simulator. One useful property of simulators is that it is typically easy to reset the environment to a previously observed state. We propose an algorithmic framework, named uncertainty-first local planning (UFLP), that takes advantage of this property. Concretely, in each data collection iteration, with some probability, our meta-algorithm resets the environment to an observed state which has high uncertainty, instead of sampling according to the initial-state distribution. The agent-environment interaction the"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2301.12579","kind":"arxiv","version":2},"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/2301.12579/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":"2301.12579","created_at":"2026-07-05T06:26:45.379057+00:00"},{"alias_kind":"arxiv_version","alias_value":"2301.12579v2","created_at":"2026-07-05T06:26:45.379057+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2301.12579","created_at":"2026-07-05T06:26:45.379057+00:00"},{"alias_kind":"pith_short_12","alias_value":"WYLLQ42RUU6M","created_at":"2026-07-05T06:26:45.379057+00:00"},{"alias_kind":"pith_short_16","alias_value":"WYLLQ42RUU6MX2WM","created_at":"2026-07-05T06:26:45.379057+00:00"},{"alias_kind":"pith_short_8","alias_value":"WYLLQ42R","created_at":"2026-07-05T06:26:45.379057+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2506.13672","citing_title":"The Courage to Stop: Overcoming Sunk Cost Fallacy in Deep Reinforcement Learning","ref_index":65,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/WYLLQ42RUU6MX2WMBQPAD6HYBM","json":"https://pith.science/pith/WYLLQ42RUU6MX2WMBQPAD6HYBM.json","graph_json":"https://pith.science/api/pith-number/WYLLQ42RUU6MX2WMBQPAD6HYBM/graph.json","events_json":"https://pith.science/api/pith-number/WYLLQ42RUU6MX2WMBQPAD6HYBM/events.json","paper":"https://pith.science/paper/WYLLQ42R"},"agent_actions":{"view_html":"https://pith.science/pith/WYLLQ42RUU6MX2WMBQPAD6HYBM","download_json":"https://pith.science/pith/WYLLQ42RUU6MX2WMBQPAD6HYBM.json","view_paper":"https://pith.science/paper/WYLLQ42R","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2301.12579&json=true","fetch_graph":"https://pith.science/api/pith-number/WYLLQ42RUU6MX2WMBQPAD6HYBM/graph.json","fetch_events":"https://pith.science/api/pith-number/WYLLQ42RUU6MX2WMBQPAD6HYBM/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/WYLLQ42RUU6MX2WMBQPAD6HYBM/action/timestamp_anchor","attest_storage":"https://pith.science/pith/WYLLQ42RUU6MX2WMBQPAD6HYBM/action/storage_attestation","attest_author":"https://pith.science/pith/WYLLQ42RUU6MX2WMBQPAD6HYBM/action/author_attestation","sign_citation":"https://pith.science/pith/WYLLQ42RUU6MX2WMBQPAD6HYBM/action/citation_signature","submit_replication":"https://pith.science/pith/WYLLQ42RUU6MX2WMBQPAD6HYBM/action/replication_record"}},"created_at":"2026-07-05T06:26:45.379057+00:00","updated_at":"2026-07-05T06:26:45.379057+00:00"}