{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:GNJLCNSBC2QMXTG4MY6BDKKPQF","short_pith_number":"pith:GNJLCNSB","schema_version":"1.0","canonical_sha256":"3352b1364116a0cbccdc663c11a94f817f41d1073c65114109036b2bc9a4468a","source":{"kind":"arxiv","id":"2409.03682","version":1},"attestation_state":"computed","paper":{"title":"A New First-Order Meta-Learning Algorithm with Convergence Guarantees","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["math.OC"],"primary_cat":"cs.LG","authors_text":"El Mahdi Chayti, Martin Jaggi","submitted_at":"2024-09-05T16:37:26Z","abstract_excerpt":"Learning new tasks by drawing on prior experience gathered from other (related) tasks is a core property of any intelligent system. Gradient-based meta-learning, especially MAML and its variants, has emerged as a viable solution to accomplish this goal. One problem MAML encounters is its computational and memory burdens needed to compute the meta-gradients. We propose a new first-order variant of MAML that we prove converges to a stationary point of the MAML objective, unlike other first-order variants. We also show that the MAML objective does not satisfy the smoothness assumption assumed in "},"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":"2409.03682","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-09-05T16:37:26Z","cross_cats_sorted":["math.OC"],"title_canon_sha256":"c50ba7d3f8a5653f5c7446c9d40371f2b225dfced8dda2ab919b6ec91bf48928","abstract_canon_sha256":"e6eb2987e3f2959d293cc19975f31d65bd0fa40c2b0871c6e8e6cebf44a0baa9"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:03:43.040877Z","signature_b64":"wsJb+qlYHMlPGFYHPfzNOr6yyX6DUhs70Hq8MF93b+ozQf4KDX96KaxjKndAugaZPn1ugW0xaS2G6Tz6kfLlDw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"3352b1364116a0cbccdc663c11a94f817f41d1073c65114109036b2bc9a4468a","last_reissued_at":"2026-07-05T09:03:43.040423Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:03:43.040423Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"A New First-Order Meta-Learning Algorithm with Convergence Guarantees","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["math.OC"],"primary_cat":"cs.LG","authors_text":"El Mahdi Chayti, Martin Jaggi","submitted_at":"2024-09-05T16:37:26Z","abstract_excerpt":"Learning new tasks by drawing on prior experience gathered from other (related) tasks is a core property of any intelligent system. Gradient-based meta-learning, especially MAML and its variants, has emerged as a viable solution to accomplish this goal. One problem MAML encounters is its computational and memory burdens needed to compute the meta-gradients. We propose a new first-order variant of MAML that we prove converges to a stationary point of the MAML objective, unlike other first-order variants. We also show that the MAML objective does not satisfy the smoothness assumption assumed in "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2409.03682","kind":"arxiv","version":1},"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/2409.03682/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":"2409.03682","created_at":"2026-07-05T09:03:43.040480+00:00"},{"alias_kind":"arxiv_version","alias_value":"2409.03682v1","created_at":"2026-07-05T09:03:43.040480+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2409.03682","created_at":"2026-07-05T09:03:43.040480+00:00"},{"alias_kind":"pith_short_12","alias_value":"GNJLCNSBC2QM","created_at":"2026-07-05T09:03:43.040480+00:00"},{"alias_kind":"pith_short_16","alias_value":"GNJLCNSBC2QMXTG4","created_at":"2026-07-05T09:03:43.040480+00:00"},{"alias_kind":"pith_short_8","alias_value":"GNJLCNSB","created_at":"2026-07-05T09:03:43.040480+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2501.06167","citing_title":"Meta-Learning for Physically-Constrained Neural System Identification","ref_index":29,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/GNJLCNSBC2QMXTG4MY6BDKKPQF","json":"https://pith.science/pith/GNJLCNSBC2QMXTG4MY6BDKKPQF.json","graph_json":"https://pith.science/api/pith-number/GNJLCNSBC2QMXTG4MY6BDKKPQF/graph.json","events_json":"https://pith.science/api/pith-number/GNJLCNSBC2QMXTG4MY6BDKKPQF/events.json","paper":"https://pith.science/paper/GNJLCNSB"},"agent_actions":{"view_html":"https://pith.science/pith/GNJLCNSBC2QMXTG4MY6BDKKPQF","download_json":"https://pith.science/pith/GNJLCNSBC2QMXTG4MY6BDKKPQF.json","view_paper":"https://pith.science/paper/GNJLCNSB","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2409.03682&json=true","fetch_graph":"https://pith.science/api/pith-number/GNJLCNSBC2QMXTG4MY6BDKKPQF/graph.json","fetch_events":"https://pith.science/api/pith-number/GNJLCNSBC2QMXTG4MY6BDKKPQF/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/GNJLCNSBC2QMXTG4MY6BDKKPQF/action/timestamp_anchor","attest_storage":"https://pith.science/pith/GNJLCNSBC2QMXTG4MY6BDKKPQF/action/storage_attestation","attest_author":"https://pith.science/pith/GNJLCNSBC2QMXTG4MY6BDKKPQF/action/author_attestation","sign_citation":"https://pith.science/pith/GNJLCNSBC2QMXTG4MY6BDKKPQF/action/citation_signature","submit_replication":"https://pith.science/pith/GNJLCNSBC2QMXTG4MY6BDKKPQF/action/replication_record"}},"created_at":"2026-07-05T09:03:43.040480+00:00","updated_at":"2026-07-05T09:03:43.040480+00:00"}