{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:5FLHODIOZN2NKE46UH423MDYEU","short_pith_number":"pith:5FLHODIO","schema_version":"1.0","canonical_sha256":"e956770d0ecb74d5139ea1f9adb078251e2dfece548177815e98f55177ab63eb","source":{"kind":"arxiv","id":"2203.14577","version":1},"attestation_state":"computed","paper":{"title":"Demystifying the Neural Tangent Kernel from a Practical Perspective: Can it be trusted for Neural Architecture Search without training?","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Byunggook Na, Dongyoon Han, Ji-Hoon Kim, Jisoo Mok, Sungroh Yoon","submitted_at":"2022-03-28T08:43:04Z","abstract_excerpt":"In Neural Architecture Search (NAS), reducing the cost of architecture evaluation remains one of the most crucial challenges. Among a plethora of efforts to bypass training of each candidate architecture to convergence for evaluation, the Neural Tangent Kernel (NTK) is emerging as a promising theoretical framework that can be utilized to estimate the performance of a neural architecture at initialization. In this work, we revisit several at-initialization metrics that can be derived from the NTK and reveal their key shortcomings. Then, through the empirical analysis of the time evolution of NT"},"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":"2203.14577","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.LG","submitted_at":"2022-03-28T08:43:04Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"f6d1f39a7127344a4b29c3132c560d7d2e3e58136940694b8aecdf583d0a2480","abstract_canon_sha256":"97d670a7a6747accd15ab133dc12af17d9d2f21985bee0e98a90013307701fff"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T04:09:06.635457Z","signature_b64":"05XUyq9WkavrIjrfz3wwKpNb8M5RZY80RHwrdgzUC6dMNjg19RSX4wVh3K0otHiNnnuNHJ4lQy+xaWxq5lPQCw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"e956770d0ecb74d5139ea1f9adb078251e2dfece548177815e98f55177ab63eb","last_reissued_at":"2026-07-05T04:09:06.635057Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T04:09:06.635057Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Demystifying the Neural Tangent Kernel from a Practical Perspective: Can it be trusted for Neural Architecture Search without training?","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Byunggook Na, Dongyoon Han, Ji-Hoon Kim, Jisoo Mok, Sungroh Yoon","submitted_at":"2022-03-28T08:43:04Z","abstract_excerpt":"In Neural Architecture Search (NAS), reducing the cost of architecture evaluation remains one of the most crucial challenges. Among a plethora of efforts to bypass training of each candidate architecture to convergence for evaluation, the Neural Tangent Kernel (NTK) is emerging as a promising theoretical framework that can be utilized to estimate the performance of a neural architecture at initialization. In this work, we revisit several at-initialization metrics that can be derived from the NTK and reveal their key shortcomings. Then, through the empirical analysis of the time evolution of NT"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2203.14577","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/2203.14577/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":"2203.14577","created_at":"2026-07-05T04:09:06.635110+00:00"},{"alias_kind":"arxiv_version","alias_value":"2203.14577v1","created_at":"2026-07-05T04:09:06.635110+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2203.14577","created_at":"2026-07-05T04:09:06.635110+00:00"},{"alias_kind":"pith_short_12","alias_value":"5FLHODIOZN2N","created_at":"2026-07-05T04:09:06.635110+00:00"},{"alias_kind":"pith_short_16","alias_value":"5FLHODIOZN2NKE46","created_at":"2026-07-05T04:09:06.635110+00:00"},{"alias_kind":"pith_short_8","alias_value":"5FLHODIO","created_at":"2026-07-05T04:09:06.635110+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.10237","citing_title":"The Benefits of Temporal Correlations: SGD Learns k-Juntas from Random Walks Efficiently","ref_index":8,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/5FLHODIOZN2NKE46UH423MDYEU","json":"https://pith.science/pith/5FLHODIOZN2NKE46UH423MDYEU.json","graph_json":"https://pith.science/api/pith-number/5FLHODIOZN2NKE46UH423MDYEU/graph.json","events_json":"https://pith.science/api/pith-number/5FLHODIOZN2NKE46UH423MDYEU/events.json","paper":"https://pith.science/paper/5FLHODIO"},"agent_actions":{"view_html":"https://pith.science/pith/5FLHODIOZN2NKE46UH423MDYEU","download_json":"https://pith.science/pith/5FLHODIOZN2NKE46UH423MDYEU.json","view_paper":"https://pith.science/paper/5FLHODIO","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2203.14577&json=true","fetch_graph":"https://pith.science/api/pith-number/5FLHODIOZN2NKE46UH423MDYEU/graph.json","fetch_events":"https://pith.science/api/pith-number/5FLHODIOZN2NKE46UH423MDYEU/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/5FLHODIOZN2NKE46UH423MDYEU/action/timestamp_anchor","attest_storage":"https://pith.science/pith/5FLHODIOZN2NKE46UH423MDYEU/action/storage_attestation","attest_author":"https://pith.science/pith/5FLHODIOZN2NKE46UH423MDYEU/action/author_attestation","sign_citation":"https://pith.science/pith/5FLHODIOZN2NKE46UH423MDYEU/action/citation_signature","submit_replication":"https://pith.science/pith/5FLHODIOZN2NKE46UH423MDYEU/action/replication_record"}},"created_at":"2026-07-05T04:09:06.635110+00:00","updated_at":"2026-07-05T04:09:06.635110+00:00"}