{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2017:3WZNGX77TOP6G7EBFMXQ77CR6P","short_pith_number":"pith:3WZNGX77","schema_version":"1.0","canonical_sha256":"ddb2d35fff9b9fe37c812b2f0ffc51f3fb8236cc06eb4b01afb5e5b4aadc8d9a","source":{"kind":"arxiv","id":"1703.08294","version":2},"attestation_state":"computed","paper":{"title":"Multi-Level Discovery of Deep Options","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Ion Stoica, Ken Goldberg, Roy Fox, Sanjay Krishnan","submitted_at":"2017-03-24T06:35:46Z","abstract_excerpt":"Augmenting an agent's control with useful higher-level behaviors called options can greatly reduce the sample complexity of reinforcement learning, but manually designing options is infeasible in high-dimensional and abstract state spaces. While recent work has proposed several techniques for automated option discovery, they do not scale to multi-level hierarchies and to expressive representations such as deep networks. We present Discovery of Deep Options (DDO), a policy-gradient algorithm that discovers parametrized options from a set of demonstration trajectories, and can be used recursivel"},"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":"1703.08294","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2017-03-24T06:35:46Z","cross_cats_sorted":[],"title_canon_sha256":"4e8cdec555cfc206aaa76e50a1b6f51b6fce65eb68b10d378fab5ff2ab7c8a2c","abstract_canon_sha256":"0f633ffaaa3c21adafaf8d52d14decc0bb9e06fcc3d13039faf0b006eb400ed8"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-18T00:33:39.317244Z","signature_b64":"9FqCsibNUV9JJ9fkgZWtim0b+KDZ08bNJjJqXLVoGJmhjiSrHEc5E4nZdmWQoOvRm+JzBTo9GiUneQVC66lUDg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"ddb2d35fff9b9fe37c812b2f0ffc51f3fb8236cc06eb4b01afb5e5b4aadc8d9a","last_reissued_at":"2026-05-18T00:33:39.316514Z","signature_status":"signed_v1","first_computed_at":"2026-05-18T00:33:39.316514Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Multi-Level Discovery of Deep Options","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Ion Stoica, Ken Goldberg, Roy Fox, Sanjay Krishnan","submitted_at":"2017-03-24T06:35:46Z","abstract_excerpt":"Augmenting an agent's control with useful higher-level behaviors called options can greatly reduce the sample complexity of reinforcement learning, but manually designing options is infeasible in high-dimensional and abstract state spaces. While recent work has proposed several techniques for automated option discovery, they do not scale to multi-level hierarchies and to expressive representations such as deep networks. We present Discovery of Deep Options (DDO), a policy-gradient algorithm that discovers parametrized options from a set of demonstration trajectories, and can be used recursivel"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1703.08294","kind":"arxiv","version":2},"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":"1703.08294","created_at":"2026-05-18T00:33:39.316638+00:00"},{"alias_kind":"arxiv_version","alias_value":"1703.08294v2","created_at":"2026-05-18T00:33:39.316638+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1703.08294","created_at":"2026-05-18T00:33:39.316638+00:00"},{"alias_kind":"pith_short_12","alias_value":"3WZNGX77TOP6","created_at":"2026-05-18T12:30:58.224056+00:00"},{"alias_kind":"pith_short_16","alias_value":"3WZNGX77TOP6G7EB","created_at":"2026-05-18T12:30:58.224056+00:00"},{"alias_kind":"pith_short_8","alias_value":"3WZNGX77","created_at":"2026-05-18T12:30:58.224056+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":2,"sample":[{"citing_arxiv_id":"1907.10993","citing_title":"Weakly Supervised Recognition of Surgical Gestures","ref_index":5,"is_internal_anchor":true},{"citing_arxiv_id":"2507.07969","citing_title":"Reinforcement Learning with Action Chunking","ref_index":18,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/3WZNGX77TOP6G7EBFMXQ77CR6P","json":"https://pith.science/pith/3WZNGX77TOP6G7EBFMXQ77CR6P.json","graph_json":"https://pith.science/api/pith-number/3WZNGX77TOP6G7EBFMXQ77CR6P/graph.json","events_json":"https://pith.science/api/pith-number/3WZNGX77TOP6G7EBFMXQ77CR6P/events.json","paper":"https://pith.science/paper/3WZNGX77"},"agent_actions":{"view_html":"https://pith.science/pith/3WZNGX77TOP6G7EBFMXQ77CR6P","download_json":"https://pith.science/pith/3WZNGX77TOP6G7EBFMXQ77CR6P.json","view_paper":"https://pith.science/paper/3WZNGX77","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1703.08294&json=true","fetch_graph":"https://pith.science/api/pith-number/3WZNGX77TOP6G7EBFMXQ77CR6P/graph.json","fetch_events":"https://pith.science/api/pith-number/3WZNGX77TOP6G7EBFMXQ77CR6P/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/3WZNGX77TOP6G7EBFMXQ77CR6P/action/timestamp_anchor","attest_storage":"https://pith.science/pith/3WZNGX77TOP6G7EBFMXQ77CR6P/action/storage_attestation","attest_author":"https://pith.science/pith/3WZNGX77TOP6G7EBFMXQ77CR6P/action/author_attestation","sign_citation":"https://pith.science/pith/3WZNGX77TOP6G7EBFMXQ77CR6P/action/citation_signature","submit_replication":"https://pith.science/pith/3WZNGX77TOP6G7EBFMXQ77CR6P/action/replication_record"}},"created_at":"2026-05-18T00:33:39.316638+00:00","updated_at":"2026-05-18T00:33:39.316638+00:00"}