{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:WQPLEK4SPP4J54G2NORZNLQLVU","short_pith_number":"pith:WQPLEK4S","schema_version":"1.0","canonical_sha256":"b41eb22b927bf89ef0da6ba396ae0bad0cb50a52ffa52fe36973aa5d1fa5326f","source":{"kind":"arxiv","id":"2506.08419","version":1},"attestation_state":"computed","paper":{"title":"Online Learning-guided Learning Rate Adaptation via Gradient Alignment","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["math.OC","stat.ML"],"primary_cat":"cs.LG","authors_text":"Ali Kavis, Aryan Mokhtari, Ruichen Jiang","submitted_at":"2025-06-10T03:46:41Z","abstract_excerpt":"The performance of an optimizer on large-scale deep learning models depends critically on fine-tuning the learning rate, often requiring an extensive grid search over base learning rates, schedules, and other hyperparameters. In this paper, we propose a principled framework called GALA (Gradient Alignment-based Learning rate Adaptation), which dynamically adjusts the learning rate by tracking the alignment between consecutive gradients and using a local curvature estimate. Guided by the convergence analysis, we formulate the problem of selecting the learning rate as a one-dimensional online le"},"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":"2506.08419","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-06-10T03:46:41Z","cross_cats_sorted":["math.OC","stat.ML"],"title_canon_sha256":"087143fbeb2c172fb897ffbeab87075f198f18beb12dc39b5293e6a2b15f2727","abstract_canon_sha256":"c6881d1b4875047e158a854a485cc55ce7e4c6188b6fd6c3f265a7260d932ad8"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:18:49.244750Z","signature_b64":"48xnHXAMnqxPPwyV2aqO7m6MfkwlMj2Bas4tn1efxCKd1YDClIK1gb8uzxf3315hjSFyKQ2BvGYMlD3Sl5pBCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"b41eb22b927bf89ef0da6ba396ae0bad0cb50a52ffa52fe36973aa5d1fa5326f","last_reissued_at":"2026-07-05T11:18:49.244354Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:18:49.244354Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Online Learning-guided Learning Rate Adaptation via Gradient Alignment","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["math.OC","stat.ML"],"primary_cat":"cs.LG","authors_text":"Ali Kavis, Aryan Mokhtari, Ruichen Jiang","submitted_at":"2025-06-10T03:46:41Z","abstract_excerpt":"The performance of an optimizer on large-scale deep learning models depends critically on fine-tuning the learning rate, often requiring an extensive grid search over base learning rates, schedules, and other hyperparameters. In this paper, we propose a principled framework called GALA (Gradient Alignment-based Learning rate Adaptation), which dynamically adjusts the learning rate by tracking the alignment between consecutive gradients and using a local curvature estimate. Guided by the convergence analysis, we formulate the problem of selecting the learning rate as a one-dimensional online le"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.08419","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/2506.08419/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":"2506.08419","created_at":"2026-07-05T11:18:49.244411+00:00"},{"alias_kind":"arxiv_version","alias_value":"2506.08419v1","created_at":"2026-07-05T11:18:49.244411+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.08419","created_at":"2026-07-05T11:18:49.244411+00:00"},{"alias_kind":"pith_short_12","alias_value":"WQPLEK4SPP4J","created_at":"2026-07-05T11:18:49.244411+00:00"},{"alias_kind":"pith_short_16","alias_value":"WQPLEK4SPP4J54G2","created_at":"2026-07-05T11:18:49.244411+00:00"},{"alias_kind":"pith_short_8","alias_value":"WQPLEK4S","created_at":"2026-07-05T11:18:49.244411+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/WQPLEK4SPP4J54G2NORZNLQLVU","json":"https://pith.science/pith/WQPLEK4SPP4J54G2NORZNLQLVU.json","graph_json":"https://pith.science/api/pith-number/WQPLEK4SPP4J54G2NORZNLQLVU/graph.json","events_json":"https://pith.science/api/pith-number/WQPLEK4SPP4J54G2NORZNLQLVU/events.json","paper":"https://pith.science/paper/WQPLEK4S"},"agent_actions":{"view_html":"https://pith.science/pith/WQPLEK4SPP4J54G2NORZNLQLVU","download_json":"https://pith.science/pith/WQPLEK4SPP4J54G2NORZNLQLVU.json","view_paper":"https://pith.science/paper/WQPLEK4S","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2506.08419&json=true","fetch_graph":"https://pith.science/api/pith-number/WQPLEK4SPP4J54G2NORZNLQLVU/graph.json","fetch_events":"https://pith.science/api/pith-number/WQPLEK4SPP4J54G2NORZNLQLVU/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/WQPLEK4SPP4J54G2NORZNLQLVU/action/timestamp_anchor","attest_storage":"https://pith.science/pith/WQPLEK4SPP4J54G2NORZNLQLVU/action/storage_attestation","attest_author":"https://pith.science/pith/WQPLEK4SPP4J54G2NORZNLQLVU/action/author_attestation","sign_citation":"https://pith.science/pith/WQPLEK4SPP4J54G2NORZNLQLVU/action/citation_signature","submit_replication":"https://pith.science/pith/WQPLEK4SPP4J54G2NORZNLQLVU/action/replication_record"}},"created_at":"2026-07-05T11:18:49.244411+00:00","updated_at":"2026-07-05T11:18:49.244411+00:00"}