{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2017:2IDMEMU2ZRXMY2JQR66CMP7CBS","short_pith_number":"pith:2IDMEMU2","schema_version":"1.0","canonical_sha256":"d206c2329acc6ecc69308fbc263fe20c89071bdc2870d79606dd2a4fbe419667","source":{"kind":"arxiv","id":"1706.08498","version":2},"attestation_state":"computed","paper":{"title":"Spectrally-normalized margin bounds for neural networks","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.NE","stat.ML"],"primary_cat":"cs.LG","authors_text":"Dylan J. Foster, Matus Telgarsky, Peter Bartlett","submitted_at":"2017-06-26T17:43:48Z","abstract_excerpt":"This paper presents a margin-based multiclass generalization bound for neural networks that scales with their margin-normalized \"spectral complexity\": their Lipschitz constant, meaning the product of the spectral norms of the weight matrices, times a certain correction factor. This bound is empirically investigated for a standard AlexNet network trained with SGD on the mnist and cifar10 datasets, with both original and random labels; the bound, the Lipschitz constants, and the excess risks are all in direct correlation, suggesting both that SGD selects predictors whose complexity scales with 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":"1706.08498","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2017-06-26T17:43:48Z","cross_cats_sorted":["cs.NE","stat.ML"],"title_canon_sha256":"52540d5254a7246d50144614babee52e157c95e5b6a9e64bada44e5784022493","abstract_canon_sha256":"444c4210dbd26364fbea772a375c61a12c7f3ab252e83b1b389687dd92bf57b4"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-18T00:28:55.272905Z","signature_b64":"tVd6QuyxnwdcdIba3GmT9/HbNQ/sDm4AOyui4yH28HJDSM+RwYK6AlX40lCfoWK269TWvmgReXlMBq0LzGuoDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"d206c2329acc6ecc69308fbc263fe20c89071bdc2870d79606dd2a4fbe419667","last_reissued_at":"2026-05-18T00:28:55.272416Z","signature_status":"signed_v1","first_computed_at":"2026-05-18T00:28:55.272416Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Spectrally-normalized margin bounds for neural networks","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.NE","stat.ML"],"primary_cat":"cs.LG","authors_text":"Dylan J. Foster, Matus Telgarsky, Peter Bartlett","submitted_at":"2017-06-26T17:43:48Z","abstract_excerpt":"This paper presents a margin-based multiclass generalization bound for neural networks that scales with their margin-normalized \"spectral complexity\": their Lipschitz constant, meaning the product of the spectral norms of the weight matrices, times a certain correction factor. This bound is empirically investigated for a standard AlexNet network trained with SGD on the mnist and cifar10 datasets, with both original and random labels; the bound, the Lipschitz constants, and the excess risks are all in direct correlation, suggesting both that SGD selects predictors whose complexity scales with t"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1706.08498","kind":"arxiv","version":2},"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":"1706.08498","created_at":"2026-05-18T00:28:55.272487+00:00"},{"alias_kind":"arxiv_version","alias_value":"1706.08498v2","created_at":"2026-05-18T00:28:55.272487+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1706.08498","created_at":"2026-05-18T00:28:55.272487+00:00"},{"alias_kind":"pith_short_12","alias_value":"2IDMEMU2ZRXM","created_at":"2026-05-18T12:30:55.937587+00:00"},{"alias_kind":"pith_short_16","alias_value":"2IDMEMU2ZRXMY2JQ","created_at":"2026-05-18T12:30:55.937587+00:00"},{"alias_kind":"pith_short_8","alias_value":"2IDMEMU2","created_at":"2026-05-18T12:30:55.937587+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2604.27723","citing_title":"Optimized Deferral for Imbalanced Settings","ref_index":4,"is_internal_anchor":false},{"citing_arxiv_id":"2605.05341","citing_title":"Feature Starvation as Geometric Instability in Sparse Autoencoders","ref_index":2,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/2IDMEMU2ZRXMY2JQR66CMP7CBS","json":"https://pith.science/pith/2IDMEMU2ZRXMY2JQR66CMP7CBS.json","graph_json":"https://pith.science/api/pith-number/2IDMEMU2ZRXMY2JQR66CMP7CBS/graph.json","events_json":"https://pith.science/api/pith-number/2IDMEMU2ZRXMY2JQR66CMP7CBS/events.json","paper":"https://pith.science/paper/2IDMEMU2"},"agent_actions":{"view_html":"https://pith.science/pith/2IDMEMU2ZRXMY2JQR66CMP7CBS","download_json":"https://pith.science/pith/2IDMEMU2ZRXMY2JQR66CMP7CBS.json","view_paper":"https://pith.science/paper/2IDMEMU2","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1706.08498&json=true","fetch_graph":"https://pith.science/api/pith-number/2IDMEMU2ZRXMY2JQR66CMP7CBS/graph.json","fetch_events":"https://pith.science/api/pith-number/2IDMEMU2ZRXMY2JQR66CMP7CBS/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/2IDMEMU2ZRXMY2JQR66CMP7CBS/action/timestamp_anchor","attest_storage":"https://pith.science/pith/2IDMEMU2ZRXMY2JQR66CMP7CBS/action/storage_attestation","attest_author":"https://pith.science/pith/2IDMEMU2ZRXMY2JQR66CMP7CBS/action/author_attestation","sign_citation":"https://pith.science/pith/2IDMEMU2ZRXMY2JQR66CMP7CBS/action/citation_signature","submit_replication":"https://pith.science/pith/2IDMEMU2ZRXMY2JQR66CMP7CBS/action/replication_record"}},"created_at":"2026-05-18T00:28:55.272487+00:00","updated_at":"2026-05-18T00:28:55.272487+00:00"}