{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:W3CGHZ5NFJP2BYZQQ4VZP6W5T3","short_pith_number":"pith:W3CGHZ5N","schema_version":"1.0","canonical_sha256":"b6c463e7ad2a5fa0e330872b97fadd9eea30b4e1e544f8cbc3dd19f479dc4187","source":{"kind":"arxiv","id":"2403.14917","version":2},"attestation_state":"computed","paper":{"title":"Mean-field Analysis on Two-layer Neural Networks from a Kernel Perspective","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Shokichi Takakura, Taiji Suzuki","submitted_at":"2024-03-22T02:41:57Z","abstract_excerpt":"In this paper, we study the feature learning ability of two-layer neural networks in the mean-field regime through the lens of kernel methods. To focus on the dynamics of the kernel induced by the first layer, we utilize a two-timescale limit, where the second layer moves much faster than the first layer. In this limit, the learning problem is reduced to the minimization problem over the intrinsic kernel. Then, we show the global convergence of the mean-field Langevin dynamics and derive time and particle discretization error. We also demonstrate that two-layer neural networks can learn a unio"},"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":"2403.14917","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2024-03-22T02:41:57Z","cross_cats_sorted":["stat.ML"],"title_canon_sha256":"50a3d0a5d11638aa9844fb6e39072ceb4a9c7c6508ab834fa6e29308e31d9f02","abstract_canon_sha256":"2971b1a471579e1c74ddb402da3b9b5072b7636a736a0bde44c67fa8a6df91ab"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:05:12.340849Z","signature_b64":"yZewKpEiu3LKhPBGikF9FmWTLVNHWp/mRKQK6Qm3SWLirPUWZBTewEkm+me8W5xgiYit1vJ/k6QbwlDjI48lDg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"b6c463e7ad2a5fa0e330872b97fadd9eea30b4e1e544f8cbc3dd19f479dc4187","last_reissued_at":"2026-07-05T08:05:12.340389Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:05:12.340389Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Mean-field Analysis on Two-layer Neural Networks from a Kernel Perspective","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Shokichi Takakura, Taiji Suzuki","submitted_at":"2024-03-22T02:41:57Z","abstract_excerpt":"In this paper, we study the feature learning ability of two-layer neural networks in the mean-field regime through the lens of kernel methods. To focus on the dynamics of the kernel induced by the first layer, we utilize a two-timescale limit, where the second layer moves much faster than the first layer. In this limit, the learning problem is reduced to the minimization problem over the intrinsic kernel. Then, we show the global convergence of the mean-field Langevin dynamics and derive time and particle discretization error. We also demonstrate that two-layer neural networks can learn a unio"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2403.14917","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/2403.14917/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":"2403.14917","created_at":"2026-07-05T08:05:12.340446+00:00"},{"alias_kind":"arxiv_version","alias_value":"2403.14917v2","created_at":"2026-07-05T08:05:12.340446+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2403.14917","created_at":"2026-07-05T08:05:12.340446+00:00"},{"alias_kind":"pith_short_12","alias_value":"W3CGHZ5NFJP2","created_at":"2026-07-05T08:05:12.340446+00:00"},{"alias_kind":"pith_short_16","alias_value":"W3CGHZ5NFJP2BYZQ","created_at":"2026-07-05T08:05:12.340446+00:00"},{"alias_kind":"pith_short_8","alias_value":"W3CGHZ5N","created_at":"2026-07-05T08:05:12.340446+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.22010","citing_title":"Uniform-in-Time Weak Propagation-of-Chaos in Shallow Neural Networks","ref_index":13,"is_internal_anchor":false},{"citing_arxiv_id":"2510.04606","citing_title":"Closed-Form Last Layer Optimization","ref_index":12,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/W3CGHZ5NFJP2BYZQQ4VZP6W5T3","json":"https://pith.science/pith/W3CGHZ5NFJP2BYZQQ4VZP6W5T3.json","graph_json":"https://pith.science/api/pith-number/W3CGHZ5NFJP2BYZQQ4VZP6W5T3/graph.json","events_json":"https://pith.science/api/pith-number/W3CGHZ5NFJP2BYZQQ4VZP6W5T3/events.json","paper":"https://pith.science/paper/W3CGHZ5N"},"agent_actions":{"view_html":"https://pith.science/pith/W3CGHZ5NFJP2BYZQQ4VZP6W5T3","download_json":"https://pith.science/pith/W3CGHZ5NFJP2BYZQQ4VZP6W5T3.json","view_paper":"https://pith.science/paper/W3CGHZ5N","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2403.14917&json=true","fetch_graph":"https://pith.science/api/pith-number/W3CGHZ5NFJP2BYZQQ4VZP6W5T3/graph.json","fetch_events":"https://pith.science/api/pith-number/W3CGHZ5NFJP2BYZQQ4VZP6W5T3/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/W3CGHZ5NFJP2BYZQQ4VZP6W5T3/action/timestamp_anchor","attest_storage":"https://pith.science/pith/W3CGHZ5NFJP2BYZQQ4VZP6W5T3/action/storage_attestation","attest_author":"https://pith.science/pith/W3CGHZ5NFJP2BYZQQ4VZP6W5T3/action/author_attestation","sign_citation":"https://pith.science/pith/W3CGHZ5NFJP2BYZQQ4VZP6W5T3/action/citation_signature","submit_replication":"https://pith.science/pith/W3CGHZ5NFJP2BYZQQ4VZP6W5T3/action/replication_record"}},"created_at":"2026-07-05T08:05:12.340446+00:00","updated_at":"2026-07-05T08:05:12.340446+00:00"}