{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:AQ6N7J3ZPYBNLGVWZNEEYBIYCC","short_pith_number":"pith:AQ6N7J3Z","schema_version":"1.0","canonical_sha256":"043cdfa7797e02d59ab6cb484c051810bd4265740644e86ae06a7f3ac26b1d45","source":{"kind":"arxiv","id":"2202.06450","version":3},"attestation_state":"computed","paper":{"title":"Towards Deployment-Efficient Reinforcement Learning: Lower Bound and Optimality","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","stat.ML"],"primary_cat":"cs.LG","authors_text":"Jiawei Huang, Jinglin Chen, Li Zhao, Nan Jiang, Tao Qin, Tie-Yan Liu","submitted_at":"2022-02-14T01:31:46Z","abstract_excerpt":"Deployment efficiency is an important criterion for many real-world applications of reinforcement learning (RL). Despite the community's increasing interest, there lacks a formal theoretical formulation for the problem. In this paper, we propose such a formulation for deployment-efficient RL (DE-RL) from an \"optimization with constraints\" perspective: we are interested in exploring an MDP and obtaining a near-optimal policy within minimal \\emph{deployment complexity}, whereas in each deployment the policy can sample a large batch of data. Using finite-horizon linear MDPs as a concrete structur"},"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":"2202.06450","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2022-02-14T01:31:46Z","cross_cats_sorted":["cs.AI","stat.ML"],"title_canon_sha256":"9dfe66374953825583ff5c2f274890f506b0a5aa5eb9c6b8b72db1a00293b0f4","abstract_canon_sha256":"8a29646179ad717f3278f10785896ed78e09c47bacd14a3cbc50eca95726f72f"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T04:53:26.902373Z","signature_b64":"Im9ojdT1qz/73PoTs7gfNp3LyviV/PdnPzDrFEBBhMNzBz8jaBl0M908sx0QA0+mD2ixHhtdA9/rmIx6ivaGCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"043cdfa7797e02d59ab6cb484c051810bd4265740644e86ae06a7f3ac26b1d45","last_reissued_at":"2026-07-05T04:53:26.901835Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T04:53:26.901835Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Towards Deployment-Efficient Reinforcement Learning: Lower Bound and Optimality","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","stat.ML"],"primary_cat":"cs.LG","authors_text":"Jiawei Huang, Jinglin Chen, Li Zhao, Nan Jiang, Tao Qin, Tie-Yan Liu","submitted_at":"2022-02-14T01:31:46Z","abstract_excerpt":"Deployment efficiency is an important criterion for many real-world applications of reinforcement learning (RL). Despite the community's increasing interest, there lacks a formal theoretical formulation for the problem. In this paper, we propose such a formulation for deployment-efficient RL (DE-RL) from an \"optimization with constraints\" perspective: we are interested in exploring an MDP and obtaining a near-optimal policy within minimal \\emph{deployment complexity}, whereas in each deployment the policy can sample a large batch of data. Using finite-horizon linear MDPs as a concrete structur"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2202.06450","kind":"arxiv","version":3},"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/2202.06450/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":"2202.06450","created_at":"2026-07-05T04:53:26.901894+00:00"},{"alias_kind":"arxiv_version","alias_value":"2202.06450v3","created_at":"2026-07-05T04:53:26.901894+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2202.06450","created_at":"2026-07-05T04:53:26.901894+00:00"},{"alias_kind":"pith_short_12","alias_value":"AQ6N7J3ZPYBN","created_at":"2026-07-05T04:53:26.901894+00:00"},{"alias_kind":"pith_short_16","alias_value":"AQ6N7J3ZPYBNLGVW","created_at":"2026-07-05T04:53:26.901894+00:00"},{"alias_kind":"pith_short_8","alias_value":"AQ6N7J3Z","created_at":"2026-07-05T04:53:26.901894+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2604.10427","citing_title":"A Queueing-Theoretic Framework for Dynamic Attack Surfaces: Data-Integrated Risk Analysis and Adaptive Defense","ref_index":30,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/AQ6N7J3ZPYBNLGVWZNEEYBIYCC","json":"https://pith.science/pith/AQ6N7J3ZPYBNLGVWZNEEYBIYCC.json","graph_json":"https://pith.science/api/pith-number/AQ6N7J3ZPYBNLGVWZNEEYBIYCC/graph.json","events_json":"https://pith.science/api/pith-number/AQ6N7J3ZPYBNLGVWZNEEYBIYCC/events.json","paper":"https://pith.science/paper/AQ6N7J3Z"},"agent_actions":{"view_html":"https://pith.science/pith/AQ6N7J3ZPYBNLGVWZNEEYBIYCC","download_json":"https://pith.science/pith/AQ6N7J3ZPYBNLGVWZNEEYBIYCC.json","view_paper":"https://pith.science/paper/AQ6N7J3Z","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2202.06450&json=true","fetch_graph":"https://pith.science/api/pith-number/AQ6N7J3ZPYBNLGVWZNEEYBIYCC/graph.json","fetch_events":"https://pith.science/api/pith-number/AQ6N7J3ZPYBNLGVWZNEEYBIYCC/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/AQ6N7J3ZPYBNLGVWZNEEYBIYCC/action/timestamp_anchor","attest_storage":"https://pith.science/pith/AQ6N7J3ZPYBNLGVWZNEEYBIYCC/action/storage_attestation","attest_author":"https://pith.science/pith/AQ6N7J3ZPYBNLGVWZNEEYBIYCC/action/author_attestation","sign_citation":"https://pith.science/pith/AQ6N7J3ZPYBNLGVWZNEEYBIYCC/action/citation_signature","submit_replication":"https://pith.science/pith/AQ6N7J3ZPYBNLGVWZNEEYBIYCC/action/replication_record"}},"created_at":"2026-07-05T04:53:26.901894+00:00","updated_at":"2026-07-05T04:53:26.901894+00:00"}