{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:HF7DD5CMV6O2HT7FGNVJEK5ORE","short_pith_number":"pith:HF7DD5CM","schema_version":"1.0","canonical_sha256":"397e31f44caf9da3cfe5336a922bae892cd262d502e317915f6f62cb80453cfa","source":{"kind":"arxiv","id":"2408.09310","version":3},"attestation_state":"computed","paper":{"title":"Narrowing the Focus: Learned Optimizers for Pretrained Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Andrey Zhmoginov, Anirudh Goyal, Gus Kristiansen, Jihwan Lee, Mark Sandler, Max Vladymyrov, Nolan Miller","submitted_at":"2024-08-17T23:55:19Z","abstract_excerpt":"In modern deep learning, the models are learned by applying gradient updates using an optimizer, which transforms the updates based on various statistics. Optimizers are often hand-designed and tuning their hyperparameters is a big part of the training process. Learned optimizers have shown some initial promise, but are generally unsuccessful as a general optimization mechanism applicable to every problem. In this work we explore a different direction: instead of learning general optimizers, we instead specialize them to a specific training environment. We propose a novel optimizer technique t"},"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":"2408.09310","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-08-17T23:55:19Z","cross_cats_sorted":[],"title_canon_sha256":"30a20c52267aa379ce1a1fb9087c356835202494c46357464eef328f38cbe080","abstract_canon_sha256":"36abe881875f0f9af6828305d9955fa1f0a35c578ff1b289977e715c861d6496"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:16:20.510206Z","signature_b64":"2odobbZ3sPsEg78XpRKEGbxOWYBx+lyZcT1ze7sQwzVuJ8FpSj8bK5VjM9ZqJRx6ZMw+IBGFWOxTcX16MfyZAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"397e31f44caf9da3cfe5336a922bae892cd262d502e317915f6f62cb80453cfa","last_reissued_at":"2026-07-05T09:16:20.509785Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:16:20.509785Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Narrowing the Focus: Learned Optimizers for Pretrained Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Andrey Zhmoginov, Anirudh Goyal, Gus Kristiansen, Jihwan Lee, Mark Sandler, Max Vladymyrov, Nolan Miller","submitted_at":"2024-08-17T23:55:19Z","abstract_excerpt":"In modern deep learning, the models are learned by applying gradient updates using an optimizer, which transforms the updates based on various statistics. Optimizers are often hand-designed and tuning their hyperparameters is a big part of the training process. Learned optimizers have shown some initial promise, but are generally unsuccessful as a general optimization mechanism applicable to every problem. In this work we explore a different direction: instead of learning general optimizers, we instead specialize them to a specific training environment. We propose a novel optimizer technique t"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2408.09310","kind":"arxiv","version":3},"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/2408.09310/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":"2408.09310","created_at":"2026-07-05T09:16:20.509841+00:00"},{"alias_kind":"arxiv_version","alias_value":"2408.09310v3","created_at":"2026-07-05T09:16:20.509841+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2408.09310","created_at":"2026-07-05T09:16:20.509841+00:00"},{"alias_kind":"pith_short_12","alias_value":"HF7DD5CMV6O2","created_at":"2026-07-05T09:16:20.509841+00:00"},{"alias_kind":"pith_short_16","alias_value":"HF7DD5CMV6O2HT7F","created_at":"2026-07-05T09:16:20.509841+00:00"},{"alias_kind":"pith_short_8","alias_value":"HF7DD5CM","created_at":"2026-07-05T09:16:20.509841+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2501.12670","citing_title":"Celo: Training Versatile Learned Optimizers on a Compute Diet","ref_index":34,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/HF7DD5CMV6O2HT7FGNVJEK5ORE","json":"https://pith.science/pith/HF7DD5CMV6O2HT7FGNVJEK5ORE.json","graph_json":"https://pith.science/api/pith-number/HF7DD5CMV6O2HT7FGNVJEK5ORE/graph.json","events_json":"https://pith.science/api/pith-number/HF7DD5CMV6O2HT7FGNVJEK5ORE/events.json","paper":"https://pith.science/paper/HF7DD5CM"},"agent_actions":{"view_html":"https://pith.science/pith/HF7DD5CMV6O2HT7FGNVJEK5ORE","download_json":"https://pith.science/pith/HF7DD5CMV6O2HT7FGNVJEK5ORE.json","view_paper":"https://pith.science/paper/HF7DD5CM","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2408.09310&json=true","fetch_graph":"https://pith.science/api/pith-number/HF7DD5CMV6O2HT7FGNVJEK5ORE/graph.json","fetch_events":"https://pith.science/api/pith-number/HF7DD5CMV6O2HT7FGNVJEK5ORE/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/HF7DD5CMV6O2HT7FGNVJEK5ORE/action/timestamp_anchor","attest_storage":"https://pith.science/pith/HF7DD5CMV6O2HT7FGNVJEK5ORE/action/storage_attestation","attest_author":"https://pith.science/pith/HF7DD5CMV6O2HT7FGNVJEK5ORE/action/author_attestation","sign_citation":"https://pith.science/pith/HF7DD5CMV6O2HT7FGNVJEK5ORE/action/citation_signature","submit_replication":"https://pith.science/pith/HF7DD5CMV6O2HT7FGNVJEK5ORE/action/replication_record"}},"created_at":"2026-07-05T09:16:20.509841+00:00","updated_at":"2026-07-05T09:16:20.509841+00:00"}