{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2018:IOY6NWHYFYRIWTBN2XQGXLRRZV","short_pith_number":"pith:IOY6NWHY","schema_version":"1.0","canonical_sha256":"43b1e6d8f82e228b4c2dd5e06bae31cd649b8b3ae29beced1c1000e88a263073","source":{"kind":"arxiv","id":"1802.01396","version":3},"attestation_state":"computed","paper":{"title":"To understand deep learning we need to understand kernel learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"stat.ML","authors_text":"Mikhail Belkin, Siyuan Ma, Soumik Mandal","submitted_at":"2018-02-05T14:14:28Z","abstract_excerpt":"Generalization performance of classifiers in deep learning has recently become a subject of intense study. Deep models, typically over-parametrized, tend to fit the training data exactly. Despite this \"overfitting\", they perform well on test data, a phenomenon not yet fully understood.\n  The first point of our paper is that strong performance of overfitted classifiers is not a unique feature of deep learning. Using six real-world and two synthetic datasets, we establish experimentally that kernel machines trained to have zero classification or near zero regression error perform very well on te"},"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":"1802.01396","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"stat.ML","submitted_at":"2018-02-05T14:14:28Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"0fbfb97416c19ea0b1344bdc72a58b7912c7475e4ded3b711e9ba0932ea0ee65","abstract_canon_sha256":"c39739fe0193fa50e622857ee4d2bdfcbda3ad387f552e0459b8461e642a440a"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-18T00:13:10.468086Z","signature_b64":"gSE5XwNrYI5gL/BXni6IORIGoZGKKESZS0/VE5wqmu4onF2MiAOpAf26vV1EFIu6SO79h7fhqDsvXY/wujnXDw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"43b1e6d8f82e228b4c2dd5e06bae31cd649b8b3ae29beced1c1000e88a263073","last_reissued_at":"2026-05-18T00:13:10.467365Z","signature_status":"signed_v1","first_computed_at":"2026-05-18T00:13:10.467365Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"To understand deep learning we need to understand kernel learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"stat.ML","authors_text":"Mikhail Belkin, Siyuan Ma, Soumik Mandal","submitted_at":"2018-02-05T14:14:28Z","abstract_excerpt":"Generalization performance of classifiers in deep learning has recently become a subject of intense study. Deep models, typically over-parametrized, tend to fit the training data exactly. Despite this \"overfitting\", they perform well on test data, a phenomenon not yet fully understood.\n  The first point of our paper is that strong performance of overfitted classifiers is not a unique feature of deep learning. Using six real-world and two synthetic datasets, we establish experimentally that kernel machines trained to have zero classification or near zero regression error perform very well on te"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1802.01396","kind":"arxiv","version":3},"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":"1802.01396","created_at":"2026-05-18T00:13:10.467483+00:00"},{"alias_kind":"arxiv_version","alias_value":"1802.01396v3","created_at":"2026-05-18T00:13:10.467483+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1802.01396","created_at":"2026-05-18T00:13:10.467483+00:00"},{"alias_kind":"pith_short_12","alias_value":"IOY6NWHYFYRI","created_at":"2026-05-18T12:32:31.084164+00:00"},{"alias_kind":"pith_short_16","alias_value":"IOY6NWHYFYRIWTBN","created_at":"2026-05-18T12:32:31.084164+00:00"},{"alias_kind":"pith_short_8","alias_value":"IOY6NWHY","created_at":"2026-05-18T12:32:31.084164+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":2,"sample":[{"citing_arxiv_id":"2606.20299","citing_title":"Statistical Properties of Training & Generalization","ref_index":160,"is_internal_anchor":true},{"citing_arxiv_id":"2606.20299","citing_title":"Statistical Properties of Training & Generalization","ref_index":160,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/IOY6NWHYFYRIWTBN2XQGXLRRZV","json":"https://pith.science/pith/IOY6NWHYFYRIWTBN2XQGXLRRZV.json","graph_json":"https://pith.science/api/pith-number/IOY6NWHYFYRIWTBN2XQGXLRRZV/graph.json","events_json":"https://pith.science/api/pith-number/IOY6NWHYFYRIWTBN2XQGXLRRZV/events.json","paper":"https://pith.science/paper/IOY6NWHY"},"agent_actions":{"view_html":"https://pith.science/pith/IOY6NWHYFYRIWTBN2XQGXLRRZV","download_json":"https://pith.science/pith/IOY6NWHYFYRIWTBN2XQGXLRRZV.json","view_paper":"https://pith.science/paper/IOY6NWHY","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1802.01396&json=true","fetch_graph":"https://pith.science/api/pith-number/IOY6NWHYFYRIWTBN2XQGXLRRZV/graph.json","fetch_events":"https://pith.science/api/pith-number/IOY6NWHYFYRIWTBN2XQGXLRRZV/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/IOY6NWHYFYRIWTBN2XQGXLRRZV/action/timestamp_anchor","attest_storage":"https://pith.science/pith/IOY6NWHYFYRIWTBN2XQGXLRRZV/action/storage_attestation","attest_author":"https://pith.science/pith/IOY6NWHYFYRIWTBN2XQGXLRRZV/action/author_attestation","sign_citation":"https://pith.science/pith/IOY6NWHYFYRIWTBN2XQGXLRRZV/action/citation_signature","submit_replication":"https://pith.science/pith/IOY6NWHYFYRIWTBN2XQGXLRRZV/action/replication_record"}},"created_at":"2026-05-18T00:13:10.467483+00:00","updated_at":"2026-05-18T00:13:10.467483+00:00"}