{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2018:QXWIRMY4UQEXUUR6PBZMBVGXFY","short_pith_number":"pith:QXWIRMY4","schema_version":"1.0","canonical_sha256":"85ec88b31ca4097a523e7872c0d4d72e21f08ff8df7d656c92e9f2d70a287385","source":{"kind":"arxiv","id":"1810.06784","version":4},"attestation_state":"computed","paper":{"title":"ProMP: Proximal Meta-Policy Search","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Dennis Lee, Ignasi Clavera, Jonas Rothfuss, Pieter Abbeel, Tamim Asfour","submitted_at":"2018-10-16T01:43:51Z","abstract_excerpt":"Credit assignment in Meta-reinforcement learning (Meta-RL) is still poorly understood. Existing methods either neglect credit assignment to pre-adaptation behavior or implement it naively. This leads to poor sample-efficiency during meta-training as well as ineffective task identification strategies. This paper provides a theoretical analysis of credit assignment in gradient-based Meta-RL. Building on the gained insights we develop a novel meta-learning algorithm that overcomes both the issue of poor credit assignment and previous difficulties in estimating meta-policy gradients. By controllin"},"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":"1810.06784","kind":"arxiv","version":4},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.LG","submitted_at":"2018-10-16T01:43:51Z","cross_cats_sorted":["stat.ML"],"title_canon_sha256":"5dfb43c7aab2c1ed0b93426dd8a6da94b3507010c1811fc41bfa6bbd0803d9df","abstract_canon_sha256":"b299d028cfab0a268b9600e9eb1761ac8d4c865b3e5dbbe4801f50a30194c84b"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T03:55:59.341744Z","signature_b64":"XEekzmbj2jyhJlkhwKFaqKjHxVQTBWSxB+f+gIR8el5B7p7+tCUddQuZaP2tykAtF2/OxKIQPtzx+tKB66g7CA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"85ec88b31ca4097a523e7872c0d4d72e21f08ff8df7d656c92e9f2d70a287385","last_reissued_at":"2026-07-05T03:55:59.341225Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T03:55:59.341225Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"ProMP: Proximal Meta-Policy Search","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Dennis Lee, Ignasi Clavera, Jonas Rothfuss, Pieter Abbeel, Tamim Asfour","submitted_at":"2018-10-16T01:43:51Z","abstract_excerpt":"Credit assignment in Meta-reinforcement learning (Meta-RL) is still poorly understood. Existing methods either neglect credit assignment to pre-adaptation behavior or implement it naively. This leads to poor sample-efficiency during meta-training as well as ineffective task identification strategies. This paper provides a theoretical analysis of credit assignment in gradient-based Meta-RL. Building on the gained insights we develop a novel meta-learning algorithm that overcomes both the issue of poor credit assignment and previous difficulties in estimating meta-policy gradients. By controllin"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1810.06784","kind":"arxiv","version":4},"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/1810.06784/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":"1810.06784","created_at":"2026-07-05T03:55:59.341283+00:00"},{"alias_kind":"arxiv_version","alias_value":"1810.06784v4","created_at":"2026-07-05T03:55:59.341283+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1810.06784","created_at":"2026-07-05T03:55:59.341283+00:00"},{"alias_kind":"pith_short_12","alias_value":"QXWIRMY4UQEX","created_at":"2026-07-05T03:55:59.341283+00:00"},{"alias_kind":"pith_short_16","alias_value":"QXWIRMY4UQEXUUR6","created_at":"2026-07-05T03:55:59.341283+00:00"},{"alias_kind":"pith_short_8","alias_value":"QXWIRMY4","created_at":"2026-07-05T03:55:59.341283+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.18315","citing_title":"Ghost Attractor Networks: Basin-Structured Dynamical Decoders for Closed-Loop Sequential Generation","ref_index":17,"is_internal_anchor":false},{"citing_arxiv_id":"2606.08533","citing_title":"Autonomous Aerial Manipulation via Contextual Contrastive Meta Reinforcement Learning","ref_index":10,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/QXWIRMY4UQEXUUR6PBZMBVGXFY","json":"https://pith.science/pith/QXWIRMY4UQEXUUR6PBZMBVGXFY.json","graph_json":"https://pith.science/api/pith-number/QXWIRMY4UQEXUUR6PBZMBVGXFY/graph.json","events_json":"https://pith.science/api/pith-number/QXWIRMY4UQEXUUR6PBZMBVGXFY/events.json","paper":"https://pith.science/paper/QXWIRMY4"},"agent_actions":{"view_html":"https://pith.science/pith/QXWIRMY4UQEXUUR6PBZMBVGXFY","download_json":"https://pith.science/pith/QXWIRMY4UQEXUUR6PBZMBVGXFY.json","view_paper":"https://pith.science/paper/QXWIRMY4","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1810.06784&json=true","fetch_graph":"https://pith.science/api/pith-number/QXWIRMY4UQEXUUR6PBZMBVGXFY/graph.json","fetch_events":"https://pith.science/api/pith-number/QXWIRMY4UQEXUUR6PBZMBVGXFY/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/QXWIRMY4UQEXUUR6PBZMBVGXFY/action/timestamp_anchor","attest_storage":"https://pith.science/pith/QXWIRMY4UQEXUUR6PBZMBVGXFY/action/storage_attestation","attest_author":"https://pith.science/pith/QXWIRMY4UQEXUUR6PBZMBVGXFY/action/author_attestation","sign_citation":"https://pith.science/pith/QXWIRMY4UQEXUUR6PBZMBVGXFY/action/citation_signature","submit_replication":"https://pith.science/pith/QXWIRMY4UQEXUUR6PBZMBVGXFY/action/replication_record"}},"created_at":"2026-07-05T03:55:59.341283+00:00","updated_at":"2026-07-05T03:55:59.341283+00:00"}