{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2016:ICACDHKPQYTJUQIFZGXINCQH6W","short_pith_number":"pith:ICACDHKP","schema_version":"1.0","canonical_sha256":"4080219d4f86269a4105c9ae868a07f5a35c250262144ba69f516778104faf65","source":{"kind":"arxiv","id":"1608.03287","version":1},"attestation_state":"computed","paper":{"title":"Deep vs. shallow networks : An approximation theory perspective","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["math.FA"],"primary_cat":"cs.LG","authors_text":"Hrushikesh Mhaskar, Tomaso Poggio","submitted_at":"2016-08-10T20:02:40Z","abstract_excerpt":"The paper briefy reviews several recent results on hierarchical architectures for learning from examples, that may formally explain the conditions under which Deep Convolutional Neural Networks perform much better in function approximation problems than shallow, one-hidden layer architectures. The paper announces new results for a non-smooth activation function - the ReLU function - used in present-day neural networks, as well as for the Gaussian networks. We propose a new definition of relative dimension to encapsulate different notions of sparsity of a function class that can possibly be exp"},"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":"1608.03287","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2016-08-10T20:02:40Z","cross_cats_sorted":["math.FA"],"title_canon_sha256":"b08689d131ba1e6ebb95b66b65505bca7752b01019338c7c028c95a9c4324ef5","abstract_canon_sha256":"34ac94afb388eb93ed12d55c3ba2f2ea9fb9bc831e757b0764d83134e35fed30"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-18T01:09:29.260093Z","signature_b64":"B4m3/POE57RUNe1+jcjk8um58dFqVxECuZSVmThkRT52scH6FQb4wzSCz2xkGNklrLjdJYb9GPc0gnMRKI0zAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"4080219d4f86269a4105c9ae868a07f5a35c250262144ba69f516778104faf65","last_reissued_at":"2026-05-18T01:09:29.259668Z","signature_status":"signed_v1","first_computed_at":"2026-05-18T01:09:29.259668Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Deep vs. shallow networks : An approximation theory perspective","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["math.FA"],"primary_cat":"cs.LG","authors_text":"Hrushikesh Mhaskar, Tomaso Poggio","submitted_at":"2016-08-10T20:02:40Z","abstract_excerpt":"The paper briefy reviews several recent results on hierarchical architectures for learning from examples, that may formally explain the conditions under which Deep Convolutional Neural Networks perform much better in function approximation problems than shallow, one-hidden layer architectures. The paper announces new results for a non-smooth activation function - the ReLU function - used in present-day neural networks, as well as for the Gaussian networks. We propose a new definition of relative dimension to encapsulate different notions of sparsity of a function class that can possibly be exp"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1608.03287","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":""},"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":"1608.03287","created_at":"2026-05-18T01:09:29.259733+00:00"},{"alias_kind":"arxiv_version","alias_value":"1608.03287v1","created_at":"2026-05-18T01:09:29.259733+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1608.03287","created_at":"2026-05-18T01:09:29.259733+00:00"},{"alias_kind":"pith_short_12","alias_value":"ICACDHKPQYTJ","created_at":"2026-05-18T12:30:22.444734+00:00"},{"alias_kind":"pith_short_16","alias_value":"ICACDHKPQYTJUQIF","created_at":"2026-05-18T12:30:22.444734+00:00"},{"alias_kind":"pith_short_8","alias_value":"ICACDHKP","created_at":"2026-05-18T12:30:22.444734+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"1611.03530","citing_title":"Understanding deep learning requires rethinking generalization","ref_index":2,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/ICACDHKPQYTJUQIFZGXINCQH6W","json":"https://pith.science/pith/ICACDHKPQYTJUQIFZGXINCQH6W.json","graph_json":"https://pith.science/api/pith-number/ICACDHKPQYTJUQIFZGXINCQH6W/graph.json","events_json":"https://pith.science/api/pith-number/ICACDHKPQYTJUQIFZGXINCQH6W/events.json","paper":"https://pith.science/paper/ICACDHKP"},"agent_actions":{"view_html":"https://pith.science/pith/ICACDHKPQYTJUQIFZGXINCQH6W","download_json":"https://pith.science/pith/ICACDHKPQYTJUQIFZGXINCQH6W.json","view_paper":"https://pith.science/paper/ICACDHKP","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1608.03287&json=true","fetch_graph":"https://pith.science/api/pith-number/ICACDHKPQYTJUQIFZGXINCQH6W/graph.json","fetch_events":"https://pith.science/api/pith-number/ICACDHKPQYTJUQIFZGXINCQH6W/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/ICACDHKPQYTJUQIFZGXINCQH6W/action/timestamp_anchor","attest_storage":"https://pith.science/pith/ICACDHKPQYTJUQIFZGXINCQH6W/action/storage_attestation","attest_author":"https://pith.science/pith/ICACDHKPQYTJUQIFZGXINCQH6W/action/author_attestation","sign_citation":"https://pith.science/pith/ICACDHKPQYTJUQIFZGXINCQH6W/action/citation_signature","submit_replication":"https://pith.science/pith/ICACDHKPQYTJUQIFZGXINCQH6W/action/replication_record"}},"created_at":"2026-05-18T01:09:29.259733+00:00","updated_at":"2026-05-18T01:09:29.259733+00:00"}