{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:Q63OFP5RLAYI5IOE7QC7BW5MCZ","short_pith_number":"pith:Q63OFP5R","schema_version":"1.0","canonical_sha256":"87b6e2bfb158308ea1c4fc05f0dbac165bc391388d751755556d60bcd5c814ed","source":{"kind":"arxiv","id":"2203.02711","version":1},"attestation_state":"computed","paper":{"title":"Meta Mirror Descent: Optimiser Learning for Fast Convergence","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":["math.OC"],"primary_cat":"cs.LG","authors_text":"Boyan Gao, Hae Beom Lee, Henry Gouk, Timothy M. Hospedales","submitted_at":"2022-03-05T11:41:13Z","abstract_excerpt":"Optimisers are an essential component for training machine learning models, and their design influences learning speed and generalisation. Several studies have attempted to learn more effective gradient-descent optimisers via solving a bi-level optimisation problem where generalisation error is minimised with respect to optimiser parameters. However, most existing optimiser learning methods are intuitively motivated, without clear theoretical support. We take a different perspective starting from mirror descent rather than gradient descent, and meta-learning the corresponding Bregman divergenc"},"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":"2203.02711","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.LG","submitted_at":"2022-03-05T11:41:13Z","cross_cats_sorted":["math.OC"],"title_canon_sha256":"dd76f8e352d2a68165f2f5afbafe411c8e9268a08873761214fa7dda0877ccd9","abstract_canon_sha256":"0741e8dcf40c1f4f479ed4f3c7eff48270b752226404e9bf135ecbba3b46aea5"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T04:02:26.363852Z","signature_b64":"zGpR+OcZtp1fyF1Qgtj46fpXRgZHi0QIO3RQRQjSagVBbfOpjXzBTiCKS4G2kpLBWeLcKc+C7YrZ2uto7D82Bg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"87b6e2bfb158308ea1c4fc05f0dbac165bc391388d751755556d60bcd5c814ed","last_reissued_at":"2026-07-05T04:02:26.363344Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T04:02:26.363344Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Meta Mirror Descent: Optimiser Learning for Fast Convergence","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":["math.OC"],"primary_cat":"cs.LG","authors_text":"Boyan Gao, Hae Beom Lee, Henry Gouk, Timothy M. Hospedales","submitted_at":"2022-03-05T11:41:13Z","abstract_excerpt":"Optimisers are an essential component for training machine learning models, and their design influences learning speed and generalisation. Several studies have attempted to learn more effective gradient-descent optimisers via solving a bi-level optimisation problem where generalisation error is minimised with respect to optimiser parameters. However, most existing optimiser learning methods are intuitively motivated, without clear theoretical support. We take a different perspective starting from mirror descent rather than gradient descent, and meta-learning the corresponding Bregman divergenc"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2203.02711","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/2203.02711/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":"2203.02711","created_at":"2026-07-05T04:02:26.363405+00:00"},{"alias_kind":"arxiv_version","alias_value":"2203.02711v1","created_at":"2026-07-05T04:02:26.363405+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2203.02711","created_at":"2026-07-05T04:02:26.363405+00:00"},{"alias_kind":"pith_short_12","alias_value":"Q63OFP5RLAYI","created_at":"2026-07-05T04:02:26.363405+00:00"},{"alias_kind":"pith_short_16","alias_value":"Q63OFP5RLAYI5IOE","created_at":"2026-07-05T04:02:26.363405+00:00"},{"alias_kind":"pith_short_8","alias_value":"Q63OFP5R","created_at":"2026-07-05T04:02:26.363405+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.28204","citing_title":"Non-Linear Strategic Classification Made Practical","ref_index":9,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/Q63OFP5RLAYI5IOE7QC7BW5MCZ","json":"https://pith.science/pith/Q63OFP5RLAYI5IOE7QC7BW5MCZ.json","graph_json":"https://pith.science/api/pith-number/Q63OFP5RLAYI5IOE7QC7BW5MCZ/graph.json","events_json":"https://pith.science/api/pith-number/Q63OFP5RLAYI5IOE7QC7BW5MCZ/events.json","paper":"https://pith.science/paper/Q63OFP5R"},"agent_actions":{"view_html":"https://pith.science/pith/Q63OFP5RLAYI5IOE7QC7BW5MCZ","download_json":"https://pith.science/pith/Q63OFP5RLAYI5IOE7QC7BW5MCZ.json","view_paper":"https://pith.science/paper/Q63OFP5R","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2203.02711&json=true","fetch_graph":"https://pith.science/api/pith-number/Q63OFP5RLAYI5IOE7QC7BW5MCZ/graph.json","fetch_events":"https://pith.science/api/pith-number/Q63OFP5RLAYI5IOE7QC7BW5MCZ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/Q63OFP5RLAYI5IOE7QC7BW5MCZ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/Q63OFP5RLAYI5IOE7QC7BW5MCZ/action/storage_attestation","attest_author":"https://pith.science/pith/Q63OFP5RLAYI5IOE7QC7BW5MCZ/action/author_attestation","sign_citation":"https://pith.science/pith/Q63OFP5RLAYI5IOE7QC7BW5MCZ/action/citation_signature","submit_replication":"https://pith.science/pith/Q63OFP5RLAYI5IOE7QC7BW5MCZ/action/replication_record"}},"created_at":"2026-07-05T04:02:26.363405+00:00","updated_at":"2026-07-05T04:02:26.363405+00:00"}