{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:QO2YHFZUQOPPKGJXQGQLEITBTN","short_pith_number":"pith:QO2YHFZU","schema_version":"1.0","canonical_sha256":"83b5839734839ef5193781a0b222619b7510bb2738895f026286214f6f3c36c6","source":{"kind":"arxiv","id":"2502.03210","version":2},"attestation_state":"computed","paper":{"title":"From Kernels to Features: A Multi-Scale Adaptive Theory of Feature Learning","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["cs.LG","stat.ML"],"primary_cat":"cond-mat.dis-nn","authors_text":"David Dahmen, Inbar Seroussi, Javed Lindner, Kirsten Fischer, Michael Kr\\\"amer, Moritz Helias, Noa Rubin, Zohar Ringel","submitted_at":"2025-02-05T14:26:50Z","abstract_excerpt":"Feature learning in neural networks is crucial for their expressive power and inductive biases, motivating various theoretical approaches. Some approaches describe network behavior after training through a change in kernel scale from initialization, resulting in a generalization power comparable to a Gaussian process. Conversely, in other approaches training results in the adaptation of the kernel to the data, involving directional changes to the kernel. The relationship and respective strengths of these two views have so far remained unresolved. This work presents a theoretical framework of m"},"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":"2502.03210","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cond-mat.dis-nn","submitted_at":"2025-02-05T14:26:50Z","cross_cats_sorted":["cs.LG","stat.ML"],"title_canon_sha256":"6f1c56ce780a1fcbbfa2312d95b4302903e3020e9d1865b6a87265abd33aacc1","abstract_canon_sha256":"3895c2f6850ce6e05501df87703d7ad892cbe99b781a84732e1734bac9ad15a0"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:10:51.429998Z","signature_b64":"+4b+egpxVYm0LMdV9X63Ww09Rl7aKaVrBJaYhYHShahyL8Y2yOj1qKtEomvKuDKSELqka3MP3QzJQst6rpgpAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"83b5839734839ef5193781a0b222619b7510bb2738895f026286214f6f3c36c6","last_reissued_at":"2026-07-05T11:10:51.429469Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:10:51.429469Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"From Kernels to Features: A Multi-Scale Adaptive Theory of Feature Learning","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["cs.LG","stat.ML"],"primary_cat":"cond-mat.dis-nn","authors_text":"David Dahmen, Inbar Seroussi, Javed Lindner, Kirsten Fischer, Michael Kr\\\"amer, Moritz Helias, Noa Rubin, Zohar Ringel","submitted_at":"2025-02-05T14:26:50Z","abstract_excerpt":"Feature learning in neural networks is crucial for their expressive power and inductive biases, motivating various theoretical approaches. Some approaches describe network behavior after training through a change in kernel scale from initialization, resulting in a generalization power comparable to a Gaussian process. Conversely, in other approaches training results in the adaptation of the kernel to the data, involving directional changes to the kernel. The relationship and respective strengths of these two views have so far remained unresolved. This work presents a theoretical framework of m"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2502.03210","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/2502.03210/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":"2502.03210","created_at":"2026-07-05T11:10:51.429528+00:00"},{"alias_kind":"arxiv_version","alias_value":"2502.03210v2","created_at":"2026-07-05T11:10:51.429528+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2502.03210","created_at":"2026-07-05T11:10:51.429528+00:00"},{"alias_kind":"pith_short_12","alias_value":"QO2YHFZUQOPP","created_at":"2026-07-05T11:10:51.429528+00:00"},{"alias_kind":"pith_short_16","alias_value":"QO2YHFZUQOPPKGJX","created_at":"2026-07-05T11:10:51.429528+00:00"},{"alias_kind":"pith_short_8","alias_value":"QO2YHFZU","created_at":"2026-07-05T11:10:51.429528+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.20299","citing_title":"Statistical Properties of Training & Generalization","ref_index":18,"is_internal_anchor":false},{"citing_arxiv_id":"2606.20299","citing_title":"Statistical Properties of Training & Generalization","ref_index":18,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/QO2YHFZUQOPPKGJXQGQLEITBTN","json":"https://pith.science/pith/QO2YHFZUQOPPKGJXQGQLEITBTN.json","graph_json":"https://pith.science/api/pith-number/QO2YHFZUQOPPKGJXQGQLEITBTN/graph.json","events_json":"https://pith.science/api/pith-number/QO2YHFZUQOPPKGJXQGQLEITBTN/events.json","paper":"https://pith.science/paper/QO2YHFZU"},"agent_actions":{"view_html":"https://pith.science/pith/QO2YHFZUQOPPKGJXQGQLEITBTN","download_json":"https://pith.science/pith/QO2YHFZUQOPPKGJXQGQLEITBTN.json","view_paper":"https://pith.science/paper/QO2YHFZU","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2502.03210&json=true","fetch_graph":"https://pith.science/api/pith-number/QO2YHFZUQOPPKGJXQGQLEITBTN/graph.json","fetch_events":"https://pith.science/api/pith-number/QO2YHFZUQOPPKGJXQGQLEITBTN/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/QO2YHFZUQOPPKGJXQGQLEITBTN/action/timestamp_anchor","attest_storage":"https://pith.science/pith/QO2YHFZUQOPPKGJXQGQLEITBTN/action/storage_attestation","attest_author":"https://pith.science/pith/QO2YHFZUQOPPKGJXQGQLEITBTN/action/author_attestation","sign_citation":"https://pith.science/pith/QO2YHFZUQOPPKGJXQGQLEITBTN/action/citation_signature","submit_replication":"https://pith.science/pith/QO2YHFZUQOPPKGJXQGQLEITBTN/action/replication_record"}},"created_at":"2026-07-05T11:10:51.429528+00:00","updated_at":"2026-07-05T11:10:51.429528+00:00"}