{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2017:5QERG2ORK35RQV22LPPOP3IGVF","short_pith_number":"pith:5QERG2OR","schema_version":"1.0","canonical_sha256":"ec091369d156fb18575a5bdee7ed06a946149524d978f2538e05451f3afa7fa9","source":{"kind":"arxiv","id":"1710.05468","version":9},"attestation_state":"computed","paper":{"title":"Generalization in Deep Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.LG","cs.NE"],"primary_cat":"stat.ML","authors_text":"Kenji Kawaguchi, Leslie Pack Kaelbling, Yoshua Bengio","submitted_at":"2017-10-16T02:21:24Z","abstract_excerpt":"This paper provides theoretical insights into why and how deep learning can generalize well, despite its large capacity, complexity, possible algorithmic instability, nonrobustness, and sharp minima, responding to an open question in the literature. We also discuss approaches to provide non-vacuous generalization guarantees for deep learning. Based on theoretical observations, we propose new open problems and discuss the limitations of our results."},"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":"1710.05468","kind":"arxiv","version":9},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"stat.ML","submitted_at":"2017-10-16T02:21:24Z","cross_cats_sorted":["cs.AI","cs.LG","cs.NE"],"title_canon_sha256":"95284b26a36cbbad80198bda24de3a62b6189a979b938058a21c9db22d1935ec","abstract_canon_sha256":"f850b8b554a0cfa068860133498f05c7fd8d406e39341f0cb0bec4a61e0452c6"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T06:45:08.364589Z","signature_b64":"Phk7txYbNOwU/wbrqnWb/c+i4jDiYjg1ZGBAjFzYeb88TvRa1XqSGG8evD/h6T19cg/uEYLqYWYrYV7+OcSHDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"ec091369d156fb18575a5bdee7ed06a946149524d978f2538e05451f3afa7fa9","last_reissued_at":"2026-07-05T06:45:08.364137Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T06:45:08.364137Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Generalization in Deep Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.LG","cs.NE"],"primary_cat":"stat.ML","authors_text":"Kenji Kawaguchi, Leslie Pack Kaelbling, Yoshua Bengio","submitted_at":"2017-10-16T02:21:24Z","abstract_excerpt":"This paper provides theoretical insights into why and how deep learning can generalize well, despite its large capacity, complexity, possible algorithmic instability, nonrobustness, and sharp minima, responding to an open question in the literature. We also discuss approaches to provide non-vacuous generalization guarantees for deep learning. Based on theoretical observations, we propose new open problems and discuss the limitations of our results."},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1710.05468","kind":"arxiv","version":9},"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/1710.05468/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":"1710.05468","created_at":"2026-07-05T06:45:08.364197+00:00"},{"alias_kind":"arxiv_version","alias_value":"1710.05468v9","created_at":"2026-07-05T06:45:08.364197+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1710.05468","created_at":"2026-07-05T06:45:08.364197+00:00"},{"alias_kind":"pith_short_12","alias_value":"5QERG2ORK35R","created_at":"2026-07-05T06:45:08.364197+00:00"},{"alias_kind":"pith_short_16","alias_value":"5QERG2ORK35RQV22","created_at":"2026-07-05T06:45:08.364197+00:00"},{"alias_kind":"pith_short_8","alias_value":"5QERG2OR","created_at":"2026-07-05T06:45:08.364197+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":7,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.28123","citing_title":"Dangerous Liaisons of Convex Learning and Non-Affine Aggregation","ref_index":69,"is_internal_anchor":false},{"citing_arxiv_id":"2505.02369","citing_title":"Sharpness-Aware Minimization with Z-Score Gradient Filtering","ref_index":15,"is_internal_anchor":false},{"citing_arxiv_id":"2511.21715","citing_title":"DNNs, Dataset Statistics, and Correlation Functions","ref_index":16,"is_internal_anchor":false},{"citing_arxiv_id":"2605.11806","citing_title":"Adaptive Kernel Ridge Regression with Linear Structure: Sharp Oracle Inequalities and Minimax Optimality","ref_index":166,"is_internal_anchor":false},{"citing_arxiv_id":"1712.00409","citing_title":"Deep Learning Scaling is Predictable, Empirically","ref_index":6,"is_internal_anchor":false},{"citing_arxiv_id":"2604.06281","citing_title":"Generalization error bounds for two-layer neural networks with Lipschitz loss function","ref_index":11,"is_internal_anchor":false},{"citing_arxiv_id":"2604.05583","citing_title":"WRF4CIR: Weight-Regularized Fine-Tuning Network for Composed Image Retrieval","ref_index":29,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/5QERG2ORK35RQV22LPPOP3IGVF","json":"https://pith.science/pith/5QERG2ORK35RQV22LPPOP3IGVF.json","graph_json":"https://pith.science/api/pith-number/5QERG2ORK35RQV22LPPOP3IGVF/graph.json","events_json":"https://pith.science/api/pith-number/5QERG2ORK35RQV22LPPOP3IGVF/events.json","paper":"https://pith.science/paper/5QERG2OR"},"agent_actions":{"view_html":"https://pith.science/pith/5QERG2ORK35RQV22LPPOP3IGVF","download_json":"https://pith.science/pith/5QERG2ORK35RQV22LPPOP3IGVF.json","view_paper":"https://pith.science/paper/5QERG2OR","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1710.05468&json=true","fetch_graph":"https://pith.science/api/pith-number/5QERG2ORK35RQV22LPPOP3IGVF/graph.json","fetch_events":"https://pith.science/api/pith-number/5QERG2ORK35RQV22LPPOP3IGVF/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/5QERG2ORK35RQV22LPPOP3IGVF/action/timestamp_anchor","attest_storage":"https://pith.science/pith/5QERG2ORK35RQV22LPPOP3IGVF/action/storage_attestation","attest_author":"https://pith.science/pith/5QERG2ORK35RQV22LPPOP3IGVF/action/author_attestation","sign_citation":"https://pith.science/pith/5QERG2ORK35RQV22LPPOP3IGVF/action/citation_signature","submit_replication":"https://pith.science/pith/5QERG2ORK35RQV22LPPOP3IGVF/action/replication_record"}},"created_at":"2026-07-05T06:45:08.364197+00:00","updated_at":"2026-07-05T06:45:08.364197+00:00"}