{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:OELRUVXTTF72HVWD7HI5IX5BS7","short_pith_number":"pith:OELRUVXT","schema_version":"1.0","canonical_sha256":"71171a56f3997fa3d6c3f9d1d45fa197f484b4782be796d06c8dcfe3d59d0efd","source":{"kind":"arxiv","id":"2306.06101","version":4},"attestation_state":"computed","paper":{"title":"Prodigy: An Expeditiously Adaptive Parameter-Free Learner","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","math.OC","stat.ML"],"primary_cat":"cs.LG","authors_text":"Aaron Defazio, Konstantin Mishchenko","submitted_at":"2023-06-09T17:59:35Z","abstract_excerpt":"We consider the problem of estimating the learning rate in adaptive methods, such as AdaGrad and Adam. We propose Prodigy, an algorithm that provably estimates the distance to the solution $D$, which is needed to set the learning rate optimally. At its core, Prodigy is a modification of the D-Adaptation method for learning-rate-free learning. It improves upon the convergence rate of D-Adaptation by a factor of $O(\\sqrt{\\log(D/d_0)})$, where $d_0$ is the initial estimate of $D$. We test Prodigy on 12 common logistic-regression benchmark datasets, VGG11 and ResNet-50 training on CIFAR10, ViT tra"},"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":"2306.06101","kind":"arxiv","version":4},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2023-06-09T17:59:35Z","cross_cats_sorted":["cs.AI","math.OC","stat.ML"],"title_canon_sha256":"fbe4e4ff4626d6693e384f6ee97c87b596a53f1603e8785ac67c5556fd4e738f","abstract_canon_sha256":"d15cc5833a0109c32298c1ffe2b4c4a05bb2edaa11b1eb7a14a6df06512930b7"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:58:20.539388Z","signature_b64":"RzKnjwUt6QIR8OtYnsDRPVpSOYAKkJSlI/qBqWemf7A+aCNpflLOiggIsJz+HPVK8WxXSP3IbTHBVJMv5yIGCw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"71171a56f3997fa3d6c3f9d1d45fa197f484b4782be796d06c8dcfe3d59d0efd","last_reissued_at":"2026-07-05T07:58:20.538883Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:58:20.538883Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Prodigy: An Expeditiously Adaptive Parameter-Free Learner","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","math.OC","stat.ML"],"primary_cat":"cs.LG","authors_text":"Aaron Defazio, Konstantin Mishchenko","submitted_at":"2023-06-09T17:59:35Z","abstract_excerpt":"We consider the problem of estimating the learning rate in adaptive methods, such as AdaGrad and Adam. We propose Prodigy, an algorithm that provably estimates the distance to the solution $D$, which is needed to set the learning rate optimally. At its core, Prodigy is a modification of the D-Adaptation method for learning-rate-free learning. It improves upon the convergence rate of D-Adaptation by a factor of $O(\\sqrt{\\log(D/d_0)})$, where $d_0$ is the initial estimate of $D$. We test Prodigy on 12 common logistic-regression benchmark datasets, VGG11 and ResNet-50 training on CIFAR10, ViT tra"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2306.06101","kind":"arxiv","version":4},"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/2306.06101/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":"2306.06101","created_at":"2026-07-05T07:58:20.538949+00:00"},{"alias_kind":"arxiv_version","alias_value":"2306.06101v4","created_at":"2026-07-05T07:58:20.538949+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2306.06101","created_at":"2026-07-05T07:58:20.538949+00:00"},{"alias_kind":"pith_short_12","alias_value":"OELRUVXTTF72","created_at":"2026-07-05T07:58:20.538949+00:00"},{"alias_kind":"pith_short_16","alias_value":"OELRUVXTTF72HVWD","created_at":"2026-07-05T07:58:20.538949+00:00"},{"alias_kind":"pith_short_8","alias_value":"OELRUVXT","created_at":"2026-07-05T07:58:20.538949+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":11,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.20556","citing_title":"Thinking in Boxes: 3D Editing in Real Images Made Easy","ref_index":39,"is_internal_anchor":false},{"citing_arxiv_id":"2606.14970","citing_title":"Zero-order Parameter-free Optimization for LMO-based Methods: Novel Approach for Efficient Fine-tuning","ref_index":56,"is_internal_anchor":false},{"citing_arxiv_id":"2606.01827","citing_title":"Adaptive Sharpness-Aware Minimization with a Polyak-type Step size: A Theory-Grounded Scheduler","ref_index":9,"is_internal_anchor":false},{"citing_arxiv_id":"2503.22171","citing_title":"An Empirical Study of Validating Synthetic Data for Text-Based Person Retrieval","ref_index":32,"is_internal_anchor":false},{"citing_arxiv_id":"2605.22644","citing_title":"Why SGD is not Brownian Motion: A New Perspective on Stochastic Dynamics","ref_index":7,"is_internal_anchor":false},{"citing_arxiv_id":"2605.20807","citing_title":"Decomposing Subject-Driven Image Generation via Intermediate Structural Prediction","ref_index":17,"is_internal_anchor":false},{"citing_arxiv_id":"2605.15388","citing_title":"Unified High-Probability Analysis of Stochastic Variance-Reduced Estimation","ref_index":101,"is_internal_anchor":false},{"citing_arxiv_id":"2409.20325","citing_title":"Old Optimizer, New Norm: An Anthology","ref_index":29,"is_internal_anchor":false},{"citing_arxiv_id":"2605.11838","citing_title":"Gradient Clipping Beyond Vector Norms: A Spectral Approach for Matrix-Valued Parameters","ref_index":42,"is_internal_anchor":false},{"citing_arxiv_id":"2505.20275","citing_title":"ImgEdit: A Unified Image Editing Dataset and Benchmark","ref_index":50,"is_internal_anchor":false},{"citing_arxiv_id":"2604.08915","citing_title":"Large-Scale Universal Defect Generation: Foundation Models and Datasets","ref_index":7,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/OELRUVXTTF72HVWD7HI5IX5BS7","json":"https://pith.science/pith/OELRUVXTTF72HVWD7HI5IX5BS7.json","graph_json":"https://pith.science/api/pith-number/OELRUVXTTF72HVWD7HI5IX5BS7/graph.json","events_json":"https://pith.science/api/pith-number/OELRUVXTTF72HVWD7HI5IX5BS7/events.json","paper":"https://pith.science/paper/OELRUVXT"},"agent_actions":{"view_html":"https://pith.science/pith/OELRUVXTTF72HVWD7HI5IX5BS7","download_json":"https://pith.science/pith/OELRUVXTTF72HVWD7HI5IX5BS7.json","view_paper":"https://pith.science/paper/OELRUVXT","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2306.06101&json=true","fetch_graph":"https://pith.science/api/pith-number/OELRUVXTTF72HVWD7HI5IX5BS7/graph.json","fetch_events":"https://pith.science/api/pith-number/OELRUVXTTF72HVWD7HI5IX5BS7/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/OELRUVXTTF72HVWD7HI5IX5BS7/action/timestamp_anchor","attest_storage":"https://pith.science/pith/OELRUVXTTF72HVWD7HI5IX5BS7/action/storage_attestation","attest_author":"https://pith.science/pith/OELRUVXTTF72HVWD7HI5IX5BS7/action/author_attestation","sign_citation":"https://pith.science/pith/OELRUVXTTF72HVWD7HI5IX5BS7/action/citation_signature","submit_replication":"https://pith.science/pith/OELRUVXTTF72HVWD7HI5IX5BS7/action/replication_record"}},"created_at":"2026-07-05T07:58:20.538949+00:00","updated_at":"2026-07-05T07:58:20.538949+00:00"}