{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2018:IKJIGZ4KFNZLIW3ITFEWDQB64V","short_pith_number":"pith:IKJIGZ4K","schema_version":"1.0","canonical_sha256":"429283678a2b72b45b68994961c03ee577ba21f84d7753c4662ee3cedd742047","source":{"kind":"arxiv","id":"1810.00113","version":2},"attestation_state":"computed","paper":{"title":"Predicting the Generalization Gap in Deep Networks with Margin Distributions","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"stat.ML","authors_text":"Dilip Krishnan, Hossein Mobahi, Samy Bengio, Yiding Jiang","submitted_at":"2018-09-28T23:23:36Z","abstract_excerpt":"As shown in recent research, deep neural networks can perfectly fit randomly labeled data, but with very poor accuracy on held out data. This phenomenon indicates that loss functions such as cross-entropy are not a reliable indicator of generalization. This leads to the crucial question of how generalization gap should be predicted from the training data and network parameters. In this paper, we propose such a measure, and conduct extensive empirical studies on how well it can predict the generalization gap. Our measure is based on the concept of margin distribution, which are the distances of"},"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":"1810.00113","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"stat.ML","submitted_at":"2018-09-28T23:23:36Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"c368986df18755ec8f1257169fee8f62d01edab0127d38bc137ee115b52a13e9","abstract_canon_sha256":"f9d370418f762b69dfae134fa761a49a2cc38282bf5d1d089954e2972f061024"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-17T23:43:33.118556Z","signature_b64":"dIG17NtMbi/bguhWDsHAojiuEm08ICsVyAXiQHjg1ZuRY/7wcIWt/YPgnfHIE6xCb7z/Ryd2AXSfg0JV/yaHDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"429283678a2b72b45b68994961c03ee577ba21f84d7753c4662ee3cedd742047","last_reissued_at":"2026-05-17T23:43:33.117774Z","signature_status":"signed_v1","first_computed_at":"2026-05-17T23:43:33.117774Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Predicting the Generalization Gap in Deep Networks with Margin Distributions","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"stat.ML","authors_text":"Dilip Krishnan, Hossein Mobahi, Samy Bengio, Yiding Jiang","submitted_at":"2018-09-28T23:23:36Z","abstract_excerpt":"As shown in recent research, deep neural networks can perfectly fit randomly labeled data, but with very poor accuracy on held out data. This phenomenon indicates that loss functions such as cross-entropy are not a reliable indicator of generalization. This leads to the crucial question of how generalization gap should be predicted from the training data and network parameters. In this paper, we propose such a measure, and conduct extensive empirical studies on how well it can predict the generalization gap. Our measure is based on the concept of margin distribution, which are the distances of"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1810.00113","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":"1810.00113","created_at":"2026-05-17T23:43:33.118019+00:00"},{"alias_kind":"arxiv_version","alias_value":"1810.00113v2","created_at":"2026-05-17T23:43:33.118019+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1810.00113","created_at":"2026-05-17T23:43:33.118019+00:00"},{"alias_kind":"pith_short_12","alias_value":"IKJIGZ4KFNZL","created_at":"2026-05-18T12:32:31.084164+00:00"},{"alias_kind":"pith_short_16","alias_value":"IKJIGZ4KFNZLIW3I","created_at":"2026-05-18T12:32:31.084164+00:00"},{"alias_kind":"pith_short_8","alias_value":"IKJIGZ4K","created_at":"2026-05-18T12:32:31.084164+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.04358","citing_title":"Intermediate Representations are Strong AI-Generated Image Detectors","ref_index":25,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/IKJIGZ4KFNZLIW3ITFEWDQB64V","json":"https://pith.science/pith/IKJIGZ4KFNZLIW3ITFEWDQB64V.json","graph_json":"https://pith.science/api/pith-number/IKJIGZ4KFNZLIW3ITFEWDQB64V/graph.json","events_json":"https://pith.science/api/pith-number/IKJIGZ4KFNZLIW3ITFEWDQB64V/events.json","paper":"https://pith.science/paper/IKJIGZ4K"},"agent_actions":{"view_html":"https://pith.science/pith/IKJIGZ4KFNZLIW3ITFEWDQB64V","download_json":"https://pith.science/pith/IKJIGZ4KFNZLIW3ITFEWDQB64V.json","view_paper":"https://pith.science/paper/IKJIGZ4K","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1810.00113&json=true","fetch_graph":"https://pith.science/api/pith-number/IKJIGZ4KFNZLIW3ITFEWDQB64V/graph.json","fetch_events":"https://pith.science/api/pith-number/IKJIGZ4KFNZLIW3ITFEWDQB64V/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/IKJIGZ4KFNZLIW3ITFEWDQB64V/action/timestamp_anchor","attest_storage":"https://pith.science/pith/IKJIGZ4KFNZLIW3ITFEWDQB64V/action/storage_attestation","attest_author":"https://pith.science/pith/IKJIGZ4KFNZLIW3ITFEWDQB64V/action/author_attestation","sign_citation":"https://pith.science/pith/IKJIGZ4KFNZLIW3ITFEWDQB64V/action/citation_signature","submit_replication":"https://pith.science/pith/IKJIGZ4KFNZLIW3ITFEWDQB64V/action/replication_record"}},"created_at":"2026-05-17T23:43:33.118019+00:00","updated_at":"2026-05-17T23:43:33.118019+00:00"}