{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:4QFJNZI4L4ZLI5WDZK2NEODJ63","short_pith_number":"pith:4QFJNZI4","schema_version":"1.0","canonical_sha256":"e40a96e51c5f32b476c3cab4d23869f6f9d3dde8687649f1709d2ee910b6c714","source":{"kind":"arxiv","id":"2306.08388","version":3},"attestation_state":"computed","paper":{"title":"Skill-Critic: Refining Learned Skills for Hierarchical Reinforcement Learning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Catherine Weaver, Ce Hao, Chen Tang, Kenta Kawamoto, Masayoshi Tomizuka, Wei Zhan","submitted_at":"2023-06-14T09:24:32Z","abstract_excerpt":"Hierarchical reinforcement learning (RL) can accelerate long-horizon decision-making by temporally abstracting a policy into multiple levels. Promising results in sparse reward environments have been seen with skills, i.e. sequences of primitive actions. Typically, a skill latent space and policy are discovered from offline data. However, the resulting low-level policy can be unreliable due to low-coverage demonstrations or distribution shifts. As a solution, we propose the Skill-Critic algorithm to fine-tune the low-level policy in conjunction with high-level skill selection. Our Skill-Critic"},"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":"2306.08388","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2023-06-14T09:24:32Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"4ead6ab98bc3847cb68073720ef050c8c4a3bf953bb3282986b19baac647a8da","abstract_canon_sha256":"d848d28e9e4d71fa3f0d458c9815631e696a738291cc39cc0fd52702120d2f23"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:42:52.883927Z","signature_b64":"rSB9TZTMOk3RsKwCuhIDKCHrhJtqlaHvP0MiBsY5ALX93kyRNnAifhLrHJCyHiqNCxvMhlZg6KqiPzm6xhwCDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"e40a96e51c5f32b476c3cab4d23869f6f9d3dde8687649f1709d2ee910b6c714","last_reissued_at":"2026-07-05T08:42:52.883560Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:42:52.883560Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Skill-Critic: Refining Learned Skills for Hierarchical Reinforcement Learning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Catherine Weaver, Ce Hao, Chen Tang, Kenta Kawamoto, Masayoshi Tomizuka, Wei Zhan","submitted_at":"2023-06-14T09:24:32Z","abstract_excerpt":"Hierarchical reinforcement learning (RL) can accelerate long-horizon decision-making by temporally abstracting a policy into multiple levels. Promising results in sparse reward environments have been seen with skills, i.e. sequences of primitive actions. Typically, a skill latent space and policy are discovered from offline data. However, the resulting low-level policy can be unreliable due to low-coverage demonstrations or distribution shifts. As a solution, we propose the Skill-Critic algorithm to fine-tune the low-level policy in conjunction with high-level skill selection. Our Skill-Critic"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2306.08388","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/2306.08388/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":"2306.08388","created_at":"2026-07-05T08:42:52.883623+00:00"},{"alias_kind":"arxiv_version","alias_value":"2306.08388v3","created_at":"2026-07-05T08:42:52.883623+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2306.08388","created_at":"2026-07-05T08:42:52.883623+00:00"},{"alias_kind":"pith_short_12","alias_value":"4QFJNZI4L4ZL","created_at":"2026-07-05T08:42:52.883623+00:00"},{"alias_kind":"pith_short_16","alias_value":"4QFJNZI4L4ZLI5WD","created_at":"2026-07-05T08:42:52.883623+00:00"},{"alias_kind":"pith_short_8","alias_value":"4QFJNZI4","created_at":"2026-07-05T08:42:52.883623+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2507.14850","citing_title":"Hierarchical Multi-Agent Reinforcement Learning with Control Barrier Functions for Safety-Critical Autonomous Systems","ref_index":32,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/4QFJNZI4L4ZLI5WDZK2NEODJ63","json":"https://pith.science/pith/4QFJNZI4L4ZLI5WDZK2NEODJ63.json","graph_json":"https://pith.science/api/pith-number/4QFJNZI4L4ZLI5WDZK2NEODJ63/graph.json","events_json":"https://pith.science/api/pith-number/4QFJNZI4L4ZLI5WDZK2NEODJ63/events.json","paper":"https://pith.science/paper/4QFJNZI4"},"agent_actions":{"view_html":"https://pith.science/pith/4QFJNZI4L4ZLI5WDZK2NEODJ63","download_json":"https://pith.science/pith/4QFJNZI4L4ZLI5WDZK2NEODJ63.json","view_paper":"https://pith.science/paper/4QFJNZI4","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2306.08388&json=true","fetch_graph":"https://pith.science/api/pith-number/4QFJNZI4L4ZLI5WDZK2NEODJ63/graph.json","fetch_events":"https://pith.science/api/pith-number/4QFJNZI4L4ZLI5WDZK2NEODJ63/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/4QFJNZI4L4ZLI5WDZK2NEODJ63/action/timestamp_anchor","attest_storage":"https://pith.science/pith/4QFJNZI4L4ZLI5WDZK2NEODJ63/action/storage_attestation","attest_author":"https://pith.science/pith/4QFJNZI4L4ZLI5WDZK2NEODJ63/action/author_attestation","sign_citation":"https://pith.science/pith/4QFJNZI4L4ZLI5WDZK2NEODJ63/action/citation_signature","submit_replication":"https://pith.science/pith/4QFJNZI4L4ZLI5WDZK2NEODJ63/action/replication_record"}},"created_at":"2026-07-05T08:42:52.883623+00:00","updated_at":"2026-07-05T08:42:52.883623+00:00"}