{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:TLHOCMWEVSPLKO7OAFAQPCUABL","short_pith_number":"pith:TLHOCMWE","schema_version":"1.0","canonical_sha256":"9acee132c4ac9eb53bee0141078a800aff19ac2f0ea86e452778ec9cd14601e3","source":{"kind":"arxiv","id":"2505.14821","version":1},"attestation_state":"computed","paper":{"title":"Sample and Computationally Efficient Continuous-Time Reinforcement Learning with General Function Approximation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Adams Yiyue Zhu, Chen Yang, Dongruo Zhou, Runze Zhao, Yue Yu","submitted_at":"2025-05-20T18:37:51Z","abstract_excerpt":"Continuous-time reinforcement learning (CTRL) provides a principled framework for sequential decision-making in environments where interactions evolve continuously over time. Despite its empirical success, the theoretical understanding of CTRL remains limited, especially in settings with general function approximation. In this work, we propose a model-based CTRL algorithm that achieves both sample and computational efficiency. Our approach leverages optimism-based confidence sets to establish the first sample complexity guarantee for CTRL with general function approximation, showing that a nea"},"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":"2505.14821","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-05-20T18:37:51Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"c449b67dab651628095ce63630042b5e3a760ecf2315f534f7cc54d03eda8925","abstract_canon_sha256":"0db50b0c71b07f9fd03686b9754aba4d58ef54153515937e3ea58a2d10fd7843"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:06:35.944692Z","signature_b64":"1wnR7ytddusC73mflV3KATtA0xQN3w6bHiCK6tpIXDtW9DITceu2k9CvtOyf7RrX4VAlgWdBUpzNKwPjkHuzBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"9acee132c4ac9eb53bee0141078a800aff19ac2f0ea86e452778ec9cd14601e3","last_reissued_at":"2026-07-05T11:06:35.944212Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:06:35.944212Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Sample and Computationally Efficient Continuous-Time Reinforcement Learning with General Function Approximation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Adams Yiyue Zhu, Chen Yang, Dongruo Zhou, Runze Zhao, Yue Yu","submitted_at":"2025-05-20T18:37:51Z","abstract_excerpt":"Continuous-time reinforcement learning (CTRL) provides a principled framework for sequential decision-making in environments where interactions evolve continuously over time. Despite its empirical success, the theoretical understanding of CTRL remains limited, especially in settings with general function approximation. In this work, we propose a model-based CTRL algorithm that achieves both sample and computational efficiency. Our approach leverages optimism-based confidence sets to establish the first sample complexity guarantee for CTRL with general function approximation, showing that a nea"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.14821","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":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2505.14821/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":"2505.14821","created_at":"2026-07-05T11:06:35.944270+00:00"},{"alias_kind":"arxiv_version","alias_value":"2505.14821v1","created_at":"2026-07-05T11:06:35.944270+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.14821","created_at":"2026-07-05T11:06:35.944270+00:00"},{"alias_kind":"pith_short_12","alias_value":"TLHOCMWEVSPL","created_at":"2026-07-05T11:06:35.944270+00:00"},{"alias_kind":"pith_short_16","alias_value":"TLHOCMWEVSPLKO7O","created_at":"2026-07-05T11:06:35.944270+00:00"},{"alias_kind":"pith_short_8","alias_value":"TLHOCMWE","created_at":"2026-07-05T11:06:35.944270+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.21925","citing_title":"PhiBE-Q-Learning: Bridging Off-Policy Reinforcement Learning and Continuous-Time Control","ref_index":46,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/TLHOCMWEVSPLKO7OAFAQPCUABL","json":"https://pith.science/pith/TLHOCMWEVSPLKO7OAFAQPCUABL.json","graph_json":"https://pith.science/api/pith-number/TLHOCMWEVSPLKO7OAFAQPCUABL/graph.json","events_json":"https://pith.science/api/pith-number/TLHOCMWEVSPLKO7OAFAQPCUABL/events.json","paper":"https://pith.science/paper/TLHOCMWE"},"agent_actions":{"view_html":"https://pith.science/pith/TLHOCMWEVSPLKO7OAFAQPCUABL","download_json":"https://pith.science/pith/TLHOCMWEVSPLKO7OAFAQPCUABL.json","view_paper":"https://pith.science/paper/TLHOCMWE","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2505.14821&json=true","fetch_graph":"https://pith.science/api/pith-number/TLHOCMWEVSPLKO7OAFAQPCUABL/graph.json","fetch_events":"https://pith.science/api/pith-number/TLHOCMWEVSPLKO7OAFAQPCUABL/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/TLHOCMWEVSPLKO7OAFAQPCUABL/action/timestamp_anchor","attest_storage":"https://pith.science/pith/TLHOCMWEVSPLKO7OAFAQPCUABL/action/storage_attestation","attest_author":"https://pith.science/pith/TLHOCMWEVSPLKO7OAFAQPCUABL/action/author_attestation","sign_citation":"https://pith.science/pith/TLHOCMWEVSPLKO7OAFAQPCUABL/action/citation_signature","submit_replication":"https://pith.science/pith/TLHOCMWEVSPLKO7OAFAQPCUABL/action/replication_record"}},"created_at":"2026-07-05T11:06:35.944270+00:00","updated_at":"2026-07-05T11:06:35.944270+00:00"}