{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:HKHAFQ4ZJWHDUM2I2LFPOCHWMY","short_pith_number":"pith:HKHAFQ4Z","schema_version":"1.0","canonical_sha256":"3a8e02c3994d8e3a3348d2caf708f6661cf4559dee535daee8c9e924da979fe8","source":{"kind":"arxiv","id":"2106.03186","version":4},"attestation_state":"computed","paper":{"title":"Reverse Engineering the Neural Tangent Kernel","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"James B. Simon, Michael R. DeWeese, Sajant Anand","submitted_at":"2021-06-06T17:25:33Z","abstract_excerpt":"The development of methods to guide the design of neural networks is an important open challenge for deep learning theory. As a paradigm for principled neural architecture design, we propose the translation of high-performing kernels, which are better-understood and amenable to first-principles design, into equivalent network architectures, which have superior efficiency, flexibility, and feature learning. To this end, we constructively prove that, with just an appropriate choice of activation function, any positive-semidefinite dot-product kernel can be realized as either the NNGP or neural t"},"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":"2106.03186","kind":"arxiv","version":4},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2021-06-06T17:25:33Z","cross_cats_sorted":[],"title_canon_sha256":"b6f802b81a275ac9b35351518c7f9e6e36147325a70bc4369ab8324fb1c3d988","abstract_canon_sha256":"85595c3dba6dfd5b32f3d70c4a804fdc796fffe6df2ef84bb272837cb860cf02"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T04:48:07.253354Z","signature_b64":"Sp4LOW6bZQWSgL69f7U2Jh5jIx/gZTmN57K3aJOS0lihiVWnFWus0LYu0bpuvqHzCIXwEbqSi+BV85qRcPViDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"3a8e02c3994d8e3a3348d2caf708f6661cf4559dee535daee8c9e924da979fe8","last_reissued_at":"2026-07-05T04:48:07.252988Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T04:48:07.252988Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Reverse Engineering the Neural Tangent Kernel","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"James B. Simon, Michael R. DeWeese, Sajant Anand","submitted_at":"2021-06-06T17:25:33Z","abstract_excerpt":"The development of methods to guide the design of neural networks is an important open challenge for deep learning theory. As a paradigm for principled neural architecture design, we propose the translation of high-performing kernels, which are better-understood and amenable to first-principles design, into equivalent network architectures, which have superior efficiency, flexibility, and feature learning. To this end, we constructively prove that, with just an appropriate choice of activation function, any positive-semidefinite dot-product kernel can be realized as either the NNGP or neural t"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2106.03186","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/2106.03186/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":"2106.03186","created_at":"2026-07-05T04:48:07.253045+00:00"},{"alias_kind":"arxiv_version","alias_value":"2106.03186v4","created_at":"2026-07-05T04:48:07.253045+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2106.03186","created_at":"2026-07-05T04:48:07.253045+00:00"},{"alias_kind":"pith_short_12","alias_value":"HKHAFQ4ZJWHD","created_at":"2026-07-05T04:48:07.253045+00:00"},{"alias_kind":"pith_short_16","alias_value":"HKHAFQ4ZJWHDUM2I","created_at":"2026-07-05T04:48:07.253045+00:00"},{"alias_kind":"pith_short_8","alias_value":"HKHAFQ4Z","created_at":"2026-07-05T04:48:07.253045+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2508.10490","citing_title":"On the Complexity-Faithfulness Trade-off of Gradient-Based Explanations","ref_index":52,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/HKHAFQ4ZJWHDUM2I2LFPOCHWMY","json":"https://pith.science/pith/HKHAFQ4ZJWHDUM2I2LFPOCHWMY.json","graph_json":"https://pith.science/api/pith-number/HKHAFQ4ZJWHDUM2I2LFPOCHWMY/graph.json","events_json":"https://pith.science/api/pith-number/HKHAFQ4ZJWHDUM2I2LFPOCHWMY/events.json","paper":"https://pith.science/paper/HKHAFQ4Z"},"agent_actions":{"view_html":"https://pith.science/pith/HKHAFQ4ZJWHDUM2I2LFPOCHWMY","download_json":"https://pith.science/pith/HKHAFQ4ZJWHDUM2I2LFPOCHWMY.json","view_paper":"https://pith.science/paper/HKHAFQ4Z","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2106.03186&json=true","fetch_graph":"https://pith.science/api/pith-number/HKHAFQ4ZJWHDUM2I2LFPOCHWMY/graph.json","fetch_events":"https://pith.science/api/pith-number/HKHAFQ4ZJWHDUM2I2LFPOCHWMY/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/HKHAFQ4ZJWHDUM2I2LFPOCHWMY/action/timestamp_anchor","attest_storage":"https://pith.science/pith/HKHAFQ4ZJWHDUM2I2LFPOCHWMY/action/storage_attestation","attest_author":"https://pith.science/pith/HKHAFQ4ZJWHDUM2I2LFPOCHWMY/action/author_attestation","sign_citation":"https://pith.science/pith/HKHAFQ4ZJWHDUM2I2LFPOCHWMY/action/citation_signature","submit_replication":"https://pith.science/pith/HKHAFQ4ZJWHDUM2I2LFPOCHWMY/action/replication_record"}},"created_at":"2026-07-05T04:48:07.253045+00:00","updated_at":"2026-07-05T04:48:07.253045+00:00"}