{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:R3YH4Z5FZDZ62Y6IAHEBQVHSD3","short_pith_number":"pith:R3YH4Z5F","schema_version":"1.0","canonical_sha256":"8ef07e67a5c8f3ed63c801c81854f21efab96fe2b545fc2e23988e11b221df35","source":{"kind":"arxiv","id":"2504.04847","version":1},"attestation_state":"computed","paper":{"title":"Nonlocal techniques for the analysis of deep ReLU neural network approximations","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CC","cs.NA","math.NA"],"primary_cat":"cs.LG","authors_text":"Cornelia Schneider, Jan Vybiral, Mario Ullrich","submitted_at":"2025-04-07T09:00:22Z","abstract_excerpt":"Recently, Daubechies, DeVore, Foucart, Hanin, and Petrova introduced a system of piece-wise linear functions, which can be easily reproduced by artificial neural networks with the ReLU activation function and which form a Riesz basis of $L_2([0,1])$. This work was generalized by two of the authors to the multivariate setting. We show that this system serves as a Riesz basis also for Sobolev spaces $W^s([0,1]^d)$ and Barron classes ${\\mathbb B}^s([0,1]^d)$ with smoothness $0<s<1$. We apply this fact to re-prove some recent results on the approximation of functions from these classes by deep neu"},"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":"2504.04847","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-04-07T09:00:22Z","cross_cats_sorted":["cs.CC","cs.NA","math.NA"],"title_canon_sha256":"ac10904e067e4263d65eaf619272047115ffdf9a818c0e10d1e41b70cfa7714f","abstract_canon_sha256":"d99ae382bcb9625ac973ff53d588ee709cc9464b003bddd401ec708e0be8c6ec"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:45:30.640264Z","signature_b64":"xghPD25S3pbW+eorxVoBxe96ktTafoR52LtFRFzZpjj03iYEh7UPX5NdeH66PgrUaYSxmiCPvhxzb+gtSGjqCw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"8ef07e67a5c8f3ed63c801c81854f21efab96fe2b545fc2e23988e11b221df35","last_reissued_at":"2026-07-05T10:45:30.639786Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:45:30.639786Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Nonlocal techniques for the analysis of deep ReLU neural network approximations","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CC","cs.NA","math.NA"],"primary_cat":"cs.LG","authors_text":"Cornelia Schneider, Jan Vybiral, Mario Ullrich","submitted_at":"2025-04-07T09:00:22Z","abstract_excerpt":"Recently, Daubechies, DeVore, Foucart, Hanin, and Petrova introduced a system of piece-wise linear functions, which can be easily reproduced by artificial neural networks with the ReLU activation function and which form a Riesz basis of $L_2([0,1])$. This work was generalized by two of the authors to the multivariate setting. We show that this system serves as a Riesz basis also for Sobolev spaces $W^s([0,1]^d)$ and Barron classes ${\\mathbb B}^s([0,1]^d)$ with smoothness $0<s<1$. We apply this fact to re-prove some recent results on the approximation of functions from these classes by deep neu"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2504.04847","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/2504.04847/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":"2504.04847","created_at":"2026-07-05T10:45:30.639847+00:00"},{"alias_kind":"arxiv_version","alias_value":"2504.04847v1","created_at":"2026-07-05T10:45:30.639847+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2504.04847","created_at":"2026-07-05T10:45:30.639847+00:00"},{"alias_kind":"pith_short_12","alias_value":"R3YH4Z5FZDZ6","created_at":"2026-07-05T10:45:30.639847+00:00"},{"alias_kind":"pith_short_16","alias_value":"R3YH4Z5FZDZ62Y6I","created_at":"2026-07-05T10:45:30.639847+00:00"},{"alias_kind":"pith_short_8","alias_value":"R3YH4Z5F","created_at":"2026-07-05T10:45:30.639847+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2509.00924","citing_title":"Beyond Universal Approximation Theorems: Algorithmic Uniform Approximation by Neural Networks Trained with Noisy Data","ref_index":98,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/R3YH4Z5FZDZ62Y6IAHEBQVHSD3","json":"https://pith.science/pith/R3YH4Z5FZDZ62Y6IAHEBQVHSD3.json","graph_json":"https://pith.science/api/pith-number/R3YH4Z5FZDZ62Y6IAHEBQVHSD3/graph.json","events_json":"https://pith.science/api/pith-number/R3YH4Z5FZDZ62Y6IAHEBQVHSD3/events.json","paper":"https://pith.science/paper/R3YH4Z5F"},"agent_actions":{"view_html":"https://pith.science/pith/R3YH4Z5FZDZ62Y6IAHEBQVHSD3","download_json":"https://pith.science/pith/R3YH4Z5FZDZ62Y6IAHEBQVHSD3.json","view_paper":"https://pith.science/paper/R3YH4Z5F","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2504.04847&json=true","fetch_graph":"https://pith.science/api/pith-number/R3YH4Z5FZDZ62Y6IAHEBQVHSD3/graph.json","fetch_events":"https://pith.science/api/pith-number/R3YH4Z5FZDZ62Y6IAHEBQVHSD3/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/R3YH4Z5FZDZ62Y6IAHEBQVHSD3/action/timestamp_anchor","attest_storage":"https://pith.science/pith/R3YH4Z5FZDZ62Y6IAHEBQVHSD3/action/storage_attestation","attest_author":"https://pith.science/pith/R3YH4Z5FZDZ62Y6IAHEBQVHSD3/action/author_attestation","sign_citation":"https://pith.science/pith/R3YH4Z5FZDZ62Y6IAHEBQVHSD3/action/citation_signature","submit_replication":"https://pith.science/pith/R3YH4Z5FZDZ62Y6IAHEBQVHSD3/action/replication_record"}},"created_at":"2026-07-05T10:45:30.639847+00:00","updated_at":"2026-07-05T10:45:30.639847+00:00"}