{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2017:7DKMRG7UHL67SZZTNYKFNEJNVR","short_pith_number":"pith:7DKMRG7U","schema_version":"1.0","canonical_sha256":"f8d4c89bf43afdf967336e1456912dac6f9a7431be1ea9ee8f727561f986f394","source":{"kind":"arxiv","id":"1705.10528","version":1},"attestation_state":"computed","paper":{"title":"Constrained Policy Optimization","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Aviv Tamar, David Held, Joshua Achiam, Pieter Abbeel","submitted_at":"2017-05-30T10:07:31Z","abstract_excerpt":"For many applications of reinforcement learning it can be more convenient to specify both a reward function and constraints, rather than trying to design behavior through the reward function. For example, systems that physically interact with or around humans should satisfy safety constraints. Recent advances in policy search algorithms (Mnih et al., 2016, Schulman et al., 2015, Lillicrap et al., 2016, Levine et al., 2016) have enabled new capabilities in high-dimensional control, but do not consider the constrained setting.\n  We propose Constrained Policy Optimization (CPO), the first general"},"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":"1705.10528","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2017-05-30T10:07:31Z","cross_cats_sorted":[],"title_canon_sha256":"dbf78666dd68bd85f88be6d02d98078ea570f23da405001333c899c89de0867e","abstract_canon_sha256":"cbcf30b550059314add3371b2a527698292cfddadf18fbe4d96f417b3409beea"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-18T00:43:24.348332Z","signature_b64":"ViOOWqJl2NeQmidWfCdjknVNXJaOfMLUyiT/2Nw6J/z5QrcoAtPfeJPiM2adbHpuh7W/+iv1EwY2Ow3CJzigCw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"f8d4c89bf43afdf967336e1456912dac6f9a7431be1ea9ee8f727561f986f394","last_reissued_at":"2026-05-18T00:43:24.347799Z","signature_status":"signed_v1","first_computed_at":"2026-05-18T00:43:24.347799Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Constrained Policy Optimization","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Aviv Tamar, David Held, Joshua Achiam, Pieter Abbeel","submitted_at":"2017-05-30T10:07:31Z","abstract_excerpt":"For many applications of reinforcement learning it can be more convenient to specify both a reward function and constraints, rather than trying to design behavior through the reward function. For example, systems that physically interact with or around humans should satisfy safety constraints. Recent advances in policy search algorithms (Mnih et al., 2016, Schulman et al., 2015, Lillicrap et al., 2016, Levine et al., 2016) have enabled new capabilities in high-dimensional control, but do not consider the constrained setting.\n  We propose Constrained Policy Optimization (CPO), the first general"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1705.10528","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":"1705.10528","created_at":"2026-05-18T00:43:24.347889+00:00"},{"alias_kind":"arxiv_version","alias_value":"1705.10528v1","created_at":"2026-05-18T00:43:24.347889+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1705.10528","created_at":"2026-05-18T00:43:24.347889+00:00"},{"alias_kind":"pith_short_12","alias_value":"7DKMRG7UHL67","created_at":"2026-05-18T12:31:03.183658+00:00"},{"alias_kind":"pith_short_16","alias_value":"7DKMRG7UHL67SZZT","created_at":"2026-05-18T12:31:03.183658+00:00"},{"alias_kind":"pith_short_8","alias_value":"7DKMRG7U","created_at":"2026-05-18T12:31:03.183658+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":5,"internal_anchor_count":3,"sample":[{"citing_arxiv_id":"1906.12189","citing_title":"Learning-based Model Predictive Control for Safe Exploration and Reinforcement Learning","ref_index":1,"is_internal_anchor":true},{"citing_arxiv_id":"2511.14135","citing_title":"AdaFair-MARL: Enforcing Adaptive Fairness Constraints in Multi-Agent Reinforcement Learning","ref_index":15,"is_internal_anchor":true},{"citing_arxiv_id":"2602.19532","citing_title":"Bellman Value Decomposition for Task Logic in Safe Optimal Control","ref_index":26,"is_internal_anchor":true},{"citing_arxiv_id":"2605.10481","citing_title":"Safe Multi-Agent Behavior Must Be Maintained, Not Merely Asserted: Constraint Drift in LLM-Based Multi-Agent Systems","ref_index":2,"is_internal_anchor":false},{"citing_arxiv_id":"2604.25848","citing_title":"Semi-Markov Reinforcement Learning for City-Scale EV Ride-Hailing with Feasibility-Guaranteed Actions","ref_index":14,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/7DKMRG7UHL67SZZTNYKFNEJNVR","json":"https://pith.science/pith/7DKMRG7UHL67SZZTNYKFNEJNVR.json","graph_json":"https://pith.science/api/pith-number/7DKMRG7UHL67SZZTNYKFNEJNVR/graph.json","events_json":"https://pith.science/api/pith-number/7DKMRG7UHL67SZZTNYKFNEJNVR/events.json","paper":"https://pith.science/paper/7DKMRG7U"},"agent_actions":{"view_html":"https://pith.science/pith/7DKMRG7UHL67SZZTNYKFNEJNVR","download_json":"https://pith.science/pith/7DKMRG7UHL67SZZTNYKFNEJNVR.json","view_paper":"https://pith.science/paper/7DKMRG7U","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1705.10528&json=true","fetch_graph":"https://pith.science/api/pith-number/7DKMRG7UHL67SZZTNYKFNEJNVR/graph.json","fetch_events":"https://pith.science/api/pith-number/7DKMRG7UHL67SZZTNYKFNEJNVR/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/7DKMRG7UHL67SZZTNYKFNEJNVR/action/timestamp_anchor","attest_storage":"https://pith.science/pith/7DKMRG7UHL67SZZTNYKFNEJNVR/action/storage_attestation","attest_author":"https://pith.science/pith/7DKMRG7UHL67SZZTNYKFNEJNVR/action/author_attestation","sign_citation":"https://pith.science/pith/7DKMRG7UHL67SZZTNYKFNEJNVR/action/citation_signature","submit_replication":"https://pith.science/pith/7DKMRG7UHL67SZZTNYKFNEJNVR/action/replication_record"}},"created_at":"2026-05-18T00:43:24.347889+00:00","updated_at":"2026-05-18T00:43:24.347889+00:00"}