{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:GKHIHSRDN3H7W6IGMKGD6SS7KG","short_pith_number":"pith:GKHIHSRD","schema_version":"1.0","canonical_sha256":"328e83ca236ecffb7906628c3f4a5f51a5030a0e388abc4eedbd9b0b7843f49c","source":{"kind":"arxiv","id":"2206.00695","version":1},"attestation_state":"computed","paper":{"title":"Know Your Boundaries: The Necessity of Explicit Behavioral Cloning in Offline RL","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Scott Niekum, Wonjoon Goo","submitted_at":"2022-06-01T18:04:43Z","abstract_excerpt":"We introduce an offline reinforcement learning (RL) algorithm that explicitly clones a behavior policy to constrain value learning. In offline RL, it is often important to prevent a policy from selecting unobserved actions, since the consequence of these actions cannot be presumed without additional information about the environment. One straightforward way to implement such a constraint is to explicitly model a given data distribution via behavior cloning and directly force a policy not to select uncertain actions. However, many offline RL methods instantiate the constraint indirectly -- for "},"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":"2206.00695","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2022-06-01T18:04:43Z","cross_cats_sorted":[],"title_canon_sha256":"38f50e2a739013331af140e573ec0b3b9d8d5a8f11ca519f203a347e76d780d8","abstract_canon_sha256":"e8686859d7ac6deb3e2a37a8ae32cfbb8ffb8c88401392cce5e123775cf4f2a9"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T04:28:32.691838Z","signature_b64":"jehjmS3rPmRs/Qo6pnmH2TWpqRL+9Zo07kjhgvGKn9jXK6zcGU6PKa0HQsZMiOvkX9SFgOROZDjhYYFBqo/MAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"328e83ca236ecffb7906628c3f4a5f51a5030a0e388abc4eedbd9b0b7843f49c","last_reissued_at":"2026-07-05T04:28:32.691352Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T04:28:32.691352Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Know Your Boundaries: The Necessity of Explicit Behavioral Cloning in Offline RL","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Scott Niekum, Wonjoon Goo","submitted_at":"2022-06-01T18:04:43Z","abstract_excerpt":"We introduce an offline reinforcement learning (RL) algorithm that explicitly clones a behavior policy to constrain value learning. In offline RL, it is often important to prevent a policy from selecting unobserved actions, since the consequence of these actions cannot be presumed without additional information about the environment. One straightforward way to implement such a constraint is to explicitly model a given data distribution via behavior cloning and directly force a policy not to select uncertain actions. However, many offline RL methods instantiate the constraint indirectly -- for "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2206.00695","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/2206.00695/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":"2206.00695","created_at":"2026-07-05T04:28:32.691411+00:00"},{"alias_kind":"arxiv_version","alias_value":"2206.00695v1","created_at":"2026-07-05T04:28:32.691411+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2206.00695","created_at":"2026-07-05T04:28:32.691411+00:00"},{"alias_kind":"pith_short_12","alias_value":"GKHIHSRDN3H7","created_at":"2026-07-05T04:28:32.691411+00:00"},{"alias_kind":"pith_short_16","alias_value":"GKHIHSRDN3H7W6IG","created_at":"2026-07-05T04:28:32.691411+00:00"},{"alias_kind":"pith_short_8","alias_value":"GKHIHSRD","created_at":"2026-07-05T04:28:32.691411+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":4,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.22630","citing_title":"Scalable Maximum Entropy Reinforcement Learning for Diffusion Policies via Adjoint Matching","ref_index":45,"is_internal_anchor":false},{"citing_arxiv_id":"2409.00588","citing_title":"Diffusion Policy Policy Optimization","ref_index":31,"is_internal_anchor":false},{"citing_arxiv_id":"2208.06193","citing_title":"Diffusion Policies as an Expressive Policy Class for Offline Reinforcement Learning","ref_index":4,"is_internal_anchor":false},{"citing_arxiv_id":"2304.10573","citing_title":"IDQL: Implicit Q-Learning as an Actor-Critic Method with Diffusion Policies","ref_index":16,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/GKHIHSRDN3H7W6IGMKGD6SS7KG","json":"https://pith.science/pith/GKHIHSRDN3H7W6IGMKGD6SS7KG.json","graph_json":"https://pith.science/api/pith-number/GKHIHSRDN3H7W6IGMKGD6SS7KG/graph.json","events_json":"https://pith.science/api/pith-number/GKHIHSRDN3H7W6IGMKGD6SS7KG/events.json","paper":"https://pith.science/paper/GKHIHSRD"},"agent_actions":{"view_html":"https://pith.science/pith/GKHIHSRDN3H7W6IGMKGD6SS7KG","download_json":"https://pith.science/pith/GKHIHSRDN3H7W6IGMKGD6SS7KG.json","view_paper":"https://pith.science/paper/GKHIHSRD","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2206.00695&json=true","fetch_graph":"https://pith.science/api/pith-number/GKHIHSRDN3H7W6IGMKGD6SS7KG/graph.json","fetch_events":"https://pith.science/api/pith-number/GKHIHSRDN3H7W6IGMKGD6SS7KG/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/GKHIHSRDN3H7W6IGMKGD6SS7KG/action/timestamp_anchor","attest_storage":"https://pith.science/pith/GKHIHSRDN3H7W6IGMKGD6SS7KG/action/storage_attestation","attest_author":"https://pith.science/pith/GKHIHSRDN3H7W6IGMKGD6SS7KG/action/author_attestation","sign_citation":"https://pith.science/pith/GKHIHSRDN3H7W6IGMKGD6SS7KG/action/citation_signature","submit_replication":"https://pith.science/pith/GKHIHSRDN3H7W6IGMKGD6SS7KG/action/replication_record"}},"created_at":"2026-07-05T04:28:32.691411+00:00","updated_at":"2026-07-05T04:28:32.691411+00:00"}