{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:DZSMDMJML4SPYKKWBKSASOQC36","short_pith_number":"pith:DZSMDMJM","schema_version":"1.0","canonical_sha256":"1e64c1b12c5f24fc29560aa4093a02dfb35380767504bc4c54a1d8fada762590","source":{"kind":"arxiv","id":"2507.17668","version":2},"attestation_state":"computed","paper":{"title":"How Should We Meta-Learn Reinforcement Learning Algorithms?","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Alexander David Goldie, Jakob Nicolaus Foerster, Jaron Cohen, Shimon Whiteson, Zilin Wang","submitted_at":"2025-07-23T16:31:38Z","abstract_excerpt":"The process of meta-learning algorithms from data, instead of relying on manual design, is growing in popularity as a paradigm for improving the performance of machine learning systems. Meta-learning shows particular promise for reinforcement learning (RL), where algorithms are often adapted from supervised or unsupervised learning despite their suboptimality for RL. However, until now there has been a severe lack of comparison between different meta-learning algorithms, such as using evolution to optimise over black-box functions or LLMs to propose code. In this paper, we carry out this empir"},"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":"2507.17668","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-07-23T16:31:38Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"5268c262f1e77f05892fae3b3e0bf0c3953e5c22f3751cc5fa8c1e9cf76557ef","abstract_canon_sha256":"24fe5ed33647d9927e2406ca0fcb833479e8385c492ce270f984e890da71ce75"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T12:08:05.565275Z","signature_b64":"qI1/WOcoToV54W43SjsCD2ru6IJ47Wynod1WxdDctrMxjUu0DLjwufKlRZbJUaT6k+luMkM5tXF3bRl9qBmMDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"1e64c1b12c5f24fc29560aa4093a02dfb35380767504bc4c54a1d8fada762590","last_reissued_at":"2026-07-05T12:08:05.564573Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T12:08:05.564573Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"How Should We Meta-Learn Reinforcement Learning Algorithms?","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Alexander David Goldie, Jakob Nicolaus Foerster, Jaron Cohen, Shimon Whiteson, Zilin Wang","submitted_at":"2025-07-23T16:31:38Z","abstract_excerpt":"The process of meta-learning algorithms from data, instead of relying on manual design, is growing in popularity as a paradigm for improving the performance of machine learning systems. Meta-learning shows particular promise for reinforcement learning (RL), where algorithms are often adapted from supervised or unsupervised learning despite their suboptimality for RL. However, until now there has been a severe lack of comparison between different meta-learning algorithms, such as using evolution to optimise over black-box functions or LLMs to propose code. In this paper, we carry out this empir"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2507.17668","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/2507.17668/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":"2507.17668","created_at":"2026-07-05T12:08:05.564673+00:00"},{"alias_kind":"arxiv_version","alias_value":"2507.17668v2","created_at":"2026-07-05T12:08:05.564673+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2507.17668","created_at":"2026-07-05T12:08:05.564673+00:00"},{"alias_kind":"pith_short_12","alias_value":"DZSMDMJML4SP","created_at":"2026-07-05T12:08:05.564673+00:00"},{"alias_kind":"pith_short_16","alias_value":"DZSMDMJML4SPYKKW","created_at":"2026-07-05T12:08:05.564673+00:00"},{"alias_kind":"pith_short_8","alias_value":"DZSMDMJM","created_at":"2026-07-05T12:08:05.564673+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2607.29559","citing_title":"LEMUR: Learning to Align with Multi-Objective Reinforcement Learning from Preference Feedback","ref_index":243,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/DZSMDMJML4SPYKKWBKSASOQC36","json":"https://pith.science/pith/DZSMDMJML4SPYKKWBKSASOQC36.json","graph_json":"https://pith.science/api/pith-number/DZSMDMJML4SPYKKWBKSASOQC36/graph.json","events_json":"https://pith.science/api/pith-number/DZSMDMJML4SPYKKWBKSASOQC36/events.json","paper":"https://pith.science/paper/DZSMDMJM"},"agent_actions":{"view_html":"https://pith.science/pith/DZSMDMJML4SPYKKWBKSASOQC36","download_json":"https://pith.science/pith/DZSMDMJML4SPYKKWBKSASOQC36.json","view_paper":"https://pith.science/paper/DZSMDMJM","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2507.17668&json=true","fetch_graph":"https://pith.science/api/pith-number/DZSMDMJML4SPYKKWBKSASOQC36/graph.json","fetch_events":"https://pith.science/api/pith-number/DZSMDMJML4SPYKKWBKSASOQC36/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/DZSMDMJML4SPYKKWBKSASOQC36/action/timestamp_anchor","attest_storage":"https://pith.science/pith/DZSMDMJML4SPYKKWBKSASOQC36/action/storage_attestation","attest_author":"https://pith.science/pith/DZSMDMJML4SPYKKWBKSASOQC36/action/author_attestation","sign_citation":"https://pith.science/pith/DZSMDMJML4SPYKKWBKSASOQC36/action/citation_signature","submit_replication":"https://pith.science/pith/DZSMDMJML4SPYKKWBKSASOQC36/action/replication_record"}},"created_at":"2026-07-05T12:08:05.564673+00:00","updated_at":"2026-07-05T12:08:05.564673+00:00"}