{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:JCMSLQ5GRPIB6EB4D2MAXCGRWL","short_pith_number":"pith:JCMSLQ5G","schema_version":"1.0","canonical_sha256":"489925c3a68bd01f103c1e980b88d1b2eb4200515662d1907562a6730849c5cb","source":{"kind":"arxiv","id":"2502.07998","version":2},"attestation_state":"computed","paper":{"title":"Adaptive kernel predictors from feature-learning infinite limits of neural networks","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cond-mat.dis-nn","stat.ML"],"primary_cat":"cs.LG","authors_text":"Blake Bordelon, Cengiz Pehlevan, Clarissa Lauditi","submitted_at":"2025-02-11T22:34:49Z","abstract_excerpt":"Previous influential work showed that infinite width limits of neural networks in the lazy training regime are described by kernel machines. Here, we show that neural networks trained in the rich, feature learning infinite-width regime in two different settings are also described by kernel machines, but with data-dependent kernels. For both cases, we provide explicit expressions for the kernel predictors and prescriptions to numerically calculate them. To derive the first predictor, we study the large-width limit of feature-learning Bayesian networks, showing how feature learning leads to task"},"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":"2502.07998","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-02-11T22:34:49Z","cross_cats_sorted":["cond-mat.dis-nn","stat.ML"],"title_canon_sha256":"e5acd49d86a6031416e14e1f1f3b25ed47aa5971533342b23696564e315777aa","abstract_canon_sha256":"e404d5f8b19b14c1dea57215165d1541668f0102100c69b826c7f008c3bb5498"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T12:08:37.379860Z","signature_b64":"wkPQYDxq8SKqAnYLW/+W+TKhxHL5AA0YrK7M3nrnHZr73nT+qeIehxXeus+jZ1IquJSKtlgzDlZMtkfwumlqBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"489925c3a68bd01f103c1e980b88d1b2eb4200515662d1907562a6730849c5cb","last_reissued_at":"2026-07-05T12:08:37.379334Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T12:08:37.379334Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Adaptive kernel predictors from feature-learning infinite limits of neural networks","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cond-mat.dis-nn","stat.ML"],"primary_cat":"cs.LG","authors_text":"Blake Bordelon, Cengiz Pehlevan, Clarissa Lauditi","submitted_at":"2025-02-11T22:34:49Z","abstract_excerpt":"Previous influential work showed that infinite width limits of neural networks in the lazy training regime are described by kernel machines. Here, we show that neural networks trained in the rich, feature learning infinite-width regime in two different settings are also described by kernel machines, but with data-dependent kernels. For both cases, we provide explicit expressions for the kernel predictors and prescriptions to numerically calculate them. To derive the first predictor, we study the large-width limit of feature-learning Bayesian networks, showing how feature learning leads to task"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2502.07998","kind":"arxiv","version":2},"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/2502.07998/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":"2502.07998","created_at":"2026-07-05T12:08:37.379396+00:00"},{"alias_kind":"arxiv_version","alias_value":"2502.07998v2","created_at":"2026-07-05T12:08:37.379396+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2502.07998","created_at":"2026-07-05T12:08:37.379396+00:00"},{"alias_kind":"pith_short_12","alias_value":"JCMSLQ5GRPIB","created_at":"2026-07-05T12:08:37.379396+00:00"},{"alias_kind":"pith_short_16","alias_value":"JCMSLQ5GRPIB6EB4","created_at":"2026-07-05T12:08:37.379396+00:00"},{"alias_kind":"pith_short_8","alias_value":"JCMSLQ5G","created_at":"2026-07-05T12:08:37.379396+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":4,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2607.05735","citing_title":"Width-Robust Learnability in Mean-Field Bayesian Neural Networks","ref_index":11,"is_internal_anchor":true},{"citing_arxiv_id":"2606.20299","citing_title":"Statistical Properties of Training & Generalization","ref_index":21,"is_internal_anchor":false},{"citing_arxiv_id":"2606.04426","citing_title":"Discrete signaling mediates chaotic regularization in recurrent neural networks","ref_index":60,"is_internal_anchor":false},{"citing_arxiv_id":"2606.20299","citing_title":"Statistical Properties of Training & Generalization","ref_index":21,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/JCMSLQ5GRPIB6EB4D2MAXCGRWL","json":"https://pith.science/pith/JCMSLQ5GRPIB6EB4D2MAXCGRWL.json","graph_json":"https://pith.science/api/pith-number/JCMSLQ5GRPIB6EB4D2MAXCGRWL/graph.json","events_json":"https://pith.science/api/pith-number/JCMSLQ5GRPIB6EB4D2MAXCGRWL/events.json","paper":"https://pith.science/paper/JCMSLQ5G"},"agent_actions":{"view_html":"https://pith.science/pith/JCMSLQ5GRPIB6EB4D2MAXCGRWL","download_json":"https://pith.science/pith/JCMSLQ5GRPIB6EB4D2MAXCGRWL.json","view_paper":"https://pith.science/paper/JCMSLQ5G","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2502.07998&json=true","fetch_graph":"https://pith.science/api/pith-number/JCMSLQ5GRPIB6EB4D2MAXCGRWL/graph.json","fetch_events":"https://pith.science/api/pith-number/JCMSLQ5GRPIB6EB4D2MAXCGRWL/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/JCMSLQ5GRPIB6EB4D2MAXCGRWL/action/timestamp_anchor","attest_storage":"https://pith.science/pith/JCMSLQ5GRPIB6EB4D2MAXCGRWL/action/storage_attestation","attest_author":"https://pith.science/pith/JCMSLQ5GRPIB6EB4D2MAXCGRWL/action/author_attestation","sign_citation":"https://pith.science/pith/JCMSLQ5GRPIB6EB4D2MAXCGRWL/action/citation_signature","submit_replication":"https://pith.science/pith/JCMSLQ5GRPIB6EB4D2MAXCGRWL/action/replication_record"}},"created_at":"2026-07-05T12:08:37.379396+00:00","updated_at":"2026-07-05T12:08:37.379396+00:00"}