{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:HNT4PAQ3VC73HFALZBKPO2LPTZ","short_pith_number":"pith:HNT4PAQ3","schema_version":"1.0","canonical_sha256":"3b67c7821ba8bfb3940bc854f7696f9e77f550a45bb8ac91ab47cbf0419190e9","source":{"kind":"arxiv","id":"2301.03236","version":1},"attestation_state":"computed","paper":{"title":"Optimistic Meta-Gradients","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","math.OC"],"primary_cat":"cs.LG","authors_text":"Andr\\'as Gy\\\"orgy, Brendan O'Donoghue, Hado van Hasselt, Satinder Singh, Sebastian Flennerhag, Tom Zahavy","submitted_at":"2023-01-09T10:05:12Z","abstract_excerpt":"We study the connection between gradient-based meta-learning and convex op-timisation. We observe that gradient descent with momentum is a special case of meta-gradients, and building on recent results in optimisation, we prove convergence rates for meta-learning in the single task setting. While a meta-learned update rule can yield faster convergence up to constant factor, it is not sufficient for acceleration. Instead, some form of optimism is required. We show that optimism in meta-learning can be captured through Bootstrapped Meta-Gradients (Flennerhag et al., 2022), providing deeper insig"},"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":"2301.03236","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2023-01-09T10:05:12Z","cross_cats_sorted":["cs.AI","math.OC"],"title_canon_sha256":"f31704e4dac86b50a86ef0df66e781fde7f4a2184cb1c1287339e709bc49208e","abstract_canon_sha256":"3b60a5329d84a4eec1ca8d79a7152d4c5721837e09beed2e1c0c77136b0233ec"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T05:31:27.024982Z","signature_b64":"QlS9dAwijYQCYLkMe7k9L3ijdN3OLxqv09YXPEscXYdunnLeG3vG3iERDSo1T14cZ4fAHnzfn2nYHXwQliI3Bw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"3b67c7821ba8bfb3940bc854f7696f9e77f550a45bb8ac91ab47cbf0419190e9","last_reissued_at":"2026-07-05T05:31:27.024560Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T05:31:27.024560Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Optimistic Meta-Gradients","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","math.OC"],"primary_cat":"cs.LG","authors_text":"Andr\\'as Gy\\\"orgy, Brendan O'Donoghue, Hado van Hasselt, Satinder Singh, Sebastian Flennerhag, Tom Zahavy","submitted_at":"2023-01-09T10:05:12Z","abstract_excerpt":"We study the connection between gradient-based meta-learning and convex op-timisation. We observe that gradient descent with momentum is a special case of meta-gradients, and building on recent results in optimisation, we prove convergence rates for meta-learning in the single task setting. While a meta-learned update rule can yield faster convergence up to constant factor, it is not sufficient for acceleration. Instead, some form of optimism is required. We show that optimism in meta-learning can be captured through Bootstrapped Meta-Gradients (Flennerhag et al., 2022), providing deeper insig"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2301.03236","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/2301.03236/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":"2301.03236","created_at":"2026-07-05T05:31:27.024622+00:00"},{"alias_kind":"arxiv_version","alias_value":"2301.03236v1","created_at":"2026-07-05T05:31:27.024622+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2301.03236","created_at":"2026-07-05T05:31:27.024622+00:00"},{"alias_kind":"pith_short_12","alias_value":"HNT4PAQ3VC73","created_at":"2026-07-05T05:31:27.024622+00:00"},{"alias_kind":"pith_short_16","alias_value":"HNT4PAQ3VC73HFAL","created_at":"2026-07-05T05:31:27.024622+00:00"},{"alias_kind":"pith_short_8","alias_value":"HNT4PAQ3","created_at":"2026-07-05T05:31:27.024622+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/HNT4PAQ3VC73HFALZBKPO2LPTZ","json":"https://pith.science/pith/HNT4PAQ3VC73HFALZBKPO2LPTZ.json","graph_json":"https://pith.science/api/pith-number/HNT4PAQ3VC73HFALZBKPO2LPTZ/graph.json","events_json":"https://pith.science/api/pith-number/HNT4PAQ3VC73HFALZBKPO2LPTZ/events.json","paper":"https://pith.science/paper/HNT4PAQ3"},"agent_actions":{"view_html":"https://pith.science/pith/HNT4PAQ3VC73HFALZBKPO2LPTZ","download_json":"https://pith.science/pith/HNT4PAQ3VC73HFALZBKPO2LPTZ.json","view_paper":"https://pith.science/paper/HNT4PAQ3","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2301.03236&json=true","fetch_graph":"https://pith.science/api/pith-number/HNT4PAQ3VC73HFALZBKPO2LPTZ/graph.json","fetch_events":"https://pith.science/api/pith-number/HNT4PAQ3VC73HFALZBKPO2LPTZ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/HNT4PAQ3VC73HFALZBKPO2LPTZ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/HNT4PAQ3VC73HFALZBKPO2LPTZ/action/storage_attestation","attest_author":"https://pith.science/pith/HNT4PAQ3VC73HFALZBKPO2LPTZ/action/author_attestation","sign_citation":"https://pith.science/pith/HNT4PAQ3VC73HFALZBKPO2LPTZ/action/citation_signature","submit_replication":"https://pith.science/pith/HNT4PAQ3VC73HFALZBKPO2LPTZ/action/replication_record"}},"created_at":"2026-07-05T05:31:27.024622+00:00","updated_at":"2026-07-05T05:31:27.024622+00:00"}