{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2018:SDGOT44FY3UXMILTLJ7R6LID2A","short_pith_number":"pith:SDGOT44F","schema_version":"1.0","canonical_sha256":"90cce9f385c6e97621735a7f1f2d03d00ee624432301f04a5c086c68fdecf760","source":{"kind":"arxiv","id":"1805.12114","version":2},"attestation_state":"computed","paper":{"title":"Deep Reinforcement Learning in a Handful of Trials using Probabilistic Dynamics Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.RO","stat.ML"],"primary_cat":"cs.LG","authors_text":"Kurtland Chua, Roberto Calandra, Rowan McAllister, Sergey Levine","submitted_at":"2018-05-30T17:55:21Z","abstract_excerpt":"Model-based reinforcement learning (RL) algorithms can attain excellent sample efficiency, but often lag behind the best model-free algorithms in terms of asymptotic performance. This is especially true with high-capacity parametric function approximators, such as deep networks. In this paper, we study how to bridge this gap, by employing uncertainty-aware dynamics models. We propose a new algorithm called probabilistic ensembles with trajectory sampling (PETS) that combines uncertainty-aware deep network dynamics models with sampling-based uncertainty propagation. Our comparison to state-of-t"},"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":"1805.12114","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2018-05-30T17:55:21Z","cross_cats_sorted":["cs.AI","cs.RO","stat.ML"],"title_canon_sha256":"cd3ca92365bb901c3039ffe3f1eb574f9ccdd285dadfdba3f13856ec36e88f2d","abstract_canon_sha256":"e84a9f0ce8324e9b0674702861b29e0d542154fb822dd183b7b8a5904eb02277"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-18T00:01:43.377025Z","signature_b64":"fa4Lo5aWYNUVEAWqVXucKK5MzUqBGlbeMLhJlVSoC4ucxrGWVgiwwbvoNQ3cAhTlBxWgjnYKH6OmWADWfzaLBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"90cce9f385c6e97621735a7f1f2d03d00ee624432301f04a5c086c68fdecf760","last_reissued_at":"2026-05-18T00:01:43.376366Z","signature_status":"signed_v1","first_computed_at":"2026-05-18T00:01:43.376366Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Deep Reinforcement Learning in a Handful of Trials using Probabilistic Dynamics Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.RO","stat.ML"],"primary_cat":"cs.LG","authors_text":"Kurtland Chua, Roberto Calandra, Rowan McAllister, Sergey Levine","submitted_at":"2018-05-30T17:55:21Z","abstract_excerpt":"Model-based reinforcement learning (RL) algorithms can attain excellent sample efficiency, but often lag behind the best model-free algorithms in terms of asymptotic performance. This is especially true with high-capacity parametric function approximators, such as deep networks. In this paper, we study how to bridge this gap, by employing uncertainty-aware dynamics models. We propose a new algorithm called probabilistic ensembles with trajectory sampling (PETS) that combines uncertainty-aware deep network dynamics models with sampling-based uncertainty propagation. Our comparison to state-of-t"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1805.12114","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":"1805.12114","created_at":"2026-05-18T00:01:43.376477+00:00"},{"alias_kind":"arxiv_version","alias_value":"1805.12114v2","created_at":"2026-05-18T00:01:43.376477+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1805.12114","created_at":"2026-05-18T00:01:43.376477+00:00"},{"alias_kind":"pith_short_12","alias_value":"SDGOT44FY3UX","created_at":"2026-05-18T12:32:53.628368+00:00"},{"alias_kind":"pith_short_16","alias_value":"SDGOT44FY3UXMILT","created_at":"2026-05-18T12:32:53.628368+00:00"},{"alias_kind":"pith_short_8","alias_value":"SDGOT44F","created_at":"2026-05-18T12:32:53.628368+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2411.04983","citing_title":"DINO-WM: World Models on Pre-trained Visual Features enable Zero-shot Planning","ref_index":10,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/SDGOT44FY3UXMILTLJ7R6LID2A","json":"https://pith.science/pith/SDGOT44FY3UXMILTLJ7R6LID2A.json","graph_json":"https://pith.science/api/pith-number/SDGOT44FY3UXMILTLJ7R6LID2A/graph.json","events_json":"https://pith.science/api/pith-number/SDGOT44FY3UXMILTLJ7R6LID2A/events.json","paper":"https://pith.science/paper/SDGOT44F"},"agent_actions":{"view_html":"https://pith.science/pith/SDGOT44FY3UXMILTLJ7R6LID2A","download_json":"https://pith.science/pith/SDGOT44FY3UXMILTLJ7R6LID2A.json","view_paper":"https://pith.science/paper/SDGOT44F","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1805.12114&json=true","fetch_graph":"https://pith.science/api/pith-number/SDGOT44FY3UXMILTLJ7R6LID2A/graph.json","fetch_events":"https://pith.science/api/pith-number/SDGOT44FY3UXMILTLJ7R6LID2A/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/SDGOT44FY3UXMILTLJ7R6LID2A/action/timestamp_anchor","attest_storage":"https://pith.science/pith/SDGOT44FY3UXMILTLJ7R6LID2A/action/storage_attestation","attest_author":"https://pith.science/pith/SDGOT44FY3UXMILTLJ7R6LID2A/action/author_attestation","sign_citation":"https://pith.science/pith/SDGOT44FY3UXMILTLJ7R6LID2A/action/citation_signature","submit_replication":"https://pith.science/pith/SDGOT44FY3UXMILTLJ7R6LID2A/action/replication_record"}},"created_at":"2026-05-18T00:01:43.376477+00:00","updated_at":"2026-05-18T00:01:43.376477+00:00"}