{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:GMZN322AMHHLP4VTKLLOTMRHUN","short_pith_number":"pith:GMZN322A","schema_version":"1.0","canonical_sha256":"3332ddeb4061ceb7f2b352d6e9b227a362af006c755ec54d4518dea0ba302b08","source":{"kind":"arxiv","id":"2106.00669","version":2},"attestation_state":"computed","paper":{"title":"Discovering Diverse Nearly Optimal Policies with Successor Features","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG","stat.ML"],"primary_cat":"cs.AI","authors_text":"Andre Barreto, Brendan O'Donoghue, Satinder Singh, Sebastian Flennerhag, Tom Zahavy, Volodymyr Mnih","submitted_at":"2021-06-01T17:56:13Z","abstract_excerpt":"Finding different solutions to the same problem is a key aspect of intelligence associated with creativity and adaptation to novel situations. In reinforcement learning, a set of diverse policies can be useful for exploration, transfer, hierarchy, and robustness. We propose Diverse Successive Policies, a method for discovering policies that are diverse in the space of Successor Features, while assuring that they are near optimal. We formalize the problem as a Constrained Markov Decision Process (CMDP) where the goal is to find policies that maximize diversity, characterized by an intrinsic div"},"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":"2106.00669","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2021-06-01T17:56:13Z","cross_cats_sorted":["cs.LG","stat.ML"],"title_canon_sha256":"f4eba4b52b9c422b11c205f960a1ba888d4823c4f3ffdaf454bf05a99d00e518","abstract_canon_sha256":"1b8909ff610623b92ebf624991a6091e2c6832b167e48187aa8097fb996ccafa"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T03:45:09.321615Z","signature_b64":"EwX4UkkB3+Dmmho4n59+v47ZFeXhO/mpxmLZ7S3SVQuHT7DFvmk3+j1QAdEI1DLr5snxuNqhkkPOlzZKlG2aAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"3332ddeb4061ceb7f2b352d6e9b227a362af006c755ec54d4518dea0ba302b08","last_reissued_at":"2026-07-05T03:45:09.321176Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T03:45:09.321176Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Discovering Diverse Nearly Optimal Policies with Successor Features","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG","stat.ML"],"primary_cat":"cs.AI","authors_text":"Andre Barreto, Brendan O'Donoghue, Satinder Singh, Sebastian Flennerhag, Tom Zahavy, Volodymyr Mnih","submitted_at":"2021-06-01T17:56:13Z","abstract_excerpt":"Finding different solutions to the same problem is a key aspect of intelligence associated with creativity and adaptation to novel situations. In reinforcement learning, a set of diverse policies can be useful for exploration, transfer, hierarchy, and robustness. We propose Diverse Successive Policies, a method for discovering policies that are diverse in the space of Successor Features, while assuring that they are near optimal. We formalize the problem as a Constrained Markov Decision Process (CMDP) where the goal is to find policies that maximize diversity, characterized by an intrinsic div"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2106.00669","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":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2106.00669/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":"2106.00669","created_at":"2026-07-05T03:45:09.321230+00:00"},{"alias_kind":"arxiv_version","alias_value":"2106.00669v2","created_at":"2026-07-05T03:45:09.321230+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2106.00669","created_at":"2026-07-05T03:45:09.321230+00:00"},{"alias_kind":"pith_short_12","alias_value":"GMZN322AMHHL","created_at":"2026-07-05T03:45:09.321230+00:00"},{"alias_kind":"pith_short_16","alias_value":"GMZN322AMHHLP4VT","created_at":"2026-07-05T03:45:09.321230+00:00"},{"alias_kind":"pith_short_8","alias_value":"GMZN322A","created_at":"2026-07-05T03:45:09.321230+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2509.01432","citing_title":"The Geometry of Nonlinear Reinforcement Learning","ref_index":2006,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/GMZN322AMHHLP4VTKLLOTMRHUN","json":"https://pith.science/pith/GMZN322AMHHLP4VTKLLOTMRHUN.json","graph_json":"https://pith.science/api/pith-number/GMZN322AMHHLP4VTKLLOTMRHUN/graph.json","events_json":"https://pith.science/api/pith-number/GMZN322AMHHLP4VTKLLOTMRHUN/events.json","paper":"https://pith.science/paper/GMZN322A"},"agent_actions":{"view_html":"https://pith.science/pith/GMZN322AMHHLP4VTKLLOTMRHUN","download_json":"https://pith.science/pith/GMZN322AMHHLP4VTKLLOTMRHUN.json","view_paper":"https://pith.science/paper/GMZN322A","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2106.00669&json=true","fetch_graph":"https://pith.science/api/pith-number/GMZN322AMHHLP4VTKLLOTMRHUN/graph.json","fetch_events":"https://pith.science/api/pith-number/GMZN322AMHHLP4VTKLLOTMRHUN/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/GMZN322AMHHLP4VTKLLOTMRHUN/action/timestamp_anchor","attest_storage":"https://pith.science/pith/GMZN322AMHHLP4VTKLLOTMRHUN/action/storage_attestation","attest_author":"https://pith.science/pith/GMZN322AMHHLP4VTKLLOTMRHUN/action/author_attestation","sign_citation":"https://pith.science/pith/GMZN322AMHHLP4VTKLLOTMRHUN/action/citation_signature","submit_replication":"https://pith.science/pith/GMZN322AMHHLP4VTKLLOTMRHUN/action/replication_record"}},"created_at":"2026-07-05T03:45:09.321230+00:00","updated_at":"2026-07-05T03:45:09.321230+00:00"}