{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2020:4F5TN5E2AD24QEOPPNGG7ULCGH","short_pith_number":"pith:4F5TN5E2","schema_version":"1.0","canonical_sha256":"e17b36f49a00f5c811cf7b4c6fd16231f54eee0e78bcee9552e173e04c9ea78b","source":{"kind":"arxiv","id":"2001.01900","version":2},"attestation_state":"computed","paper":{"title":"Regularization via Structural Label Smoothing","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Gautam Dasarathy, Visar Berisha, Weizhi Li","submitted_at":"2020-01-07T05:45:18Z","abstract_excerpt":"Regularization is an effective way to promote the generalization performance of machine learning models. In this paper, we focus on label smoothing, a form of output distribution regularization that prevents overfitting of a neural network by softening the ground-truth labels in the training data in an attempt to penalize overconfident outputs. Existing approaches typically use cross-validation to impose this smoothing, which is uniform across all training data. In this paper, we show that such label smoothing imposes a quantifiable bias in the Bayes error rate of the training data, with regio"},"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":"2001.01900","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2020-01-07T05:45:18Z","cross_cats_sorted":["stat.ML"],"title_canon_sha256":"9de40cb93c3d2902d2b73b4f78bef0e8a690661b3311a97b2aead4b49f791b6f","abstract_canon_sha256":"9fab65f66d4439717ea230b3f20e93d744d97f7bd3b2e19e279f7873b3338b45"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T01:16:11.766490Z","signature_b64":"7uPnw8PvM7bADSIGbJkvCY4tRb5dyldTOSysiTUXNAsqCqVk7IFkQIq/hsOFlJlcUQsqepTkeukxB7RTRc04DA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"e17b36f49a00f5c811cf7b4c6fd16231f54eee0e78bcee9552e173e04c9ea78b","last_reissued_at":"2026-07-05T01:16:11.765921Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T01:16:11.765921Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Regularization via Structural Label Smoothing","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Gautam Dasarathy, Visar Berisha, Weizhi Li","submitted_at":"2020-01-07T05:45:18Z","abstract_excerpt":"Regularization is an effective way to promote the generalization performance of machine learning models. In this paper, we focus on label smoothing, a form of output distribution regularization that prevents overfitting of a neural network by softening the ground-truth labels in the training data in an attempt to penalize overconfident outputs. Existing approaches typically use cross-validation to impose this smoothing, which is uniform across all training data. In this paper, we show that such label smoothing imposes a quantifiable bias in the Bayes error rate of the training data, with regio"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2001.01900","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":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2001.01900/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":"2001.01900","created_at":"2026-07-05T01:16:11.766017+00:00"},{"alias_kind":"arxiv_version","alias_value":"2001.01900v2","created_at":"2026-07-05T01:16:11.766017+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2001.01900","created_at":"2026-07-05T01:16:11.766017+00:00"},{"alias_kind":"pith_short_12","alias_value":"4F5TN5E2AD24","created_at":"2026-07-05T01:16:11.766017+00:00"},{"alias_kind":"pith_short_16","alias_value":"4F5TN5E2AD24QEOP","created_at":"2026-07-05T01:16:11.766017+00:00"},{"alias_kind":"pith_short_8","alias_value":"4F5TN5E2","created_at":"2026-07-05T01:16:11.766017+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/4F5TN5E2AD24QEOPPNGG7ULCGH","json":"https://pith.science/pith/4F5TN5E2AD24QEOPPNGG7ULCGH.json","graph_json":"https://pith.science/api/pith-number/4F5TN5E2AD24QEOPPNGG7ULCGH/graph.json","events_json":"https://pith.science/api/pith-number/4F5TN5E2AD24QEOPPNGG7ULCGH/events.json","paper":"https://pith.science/paper/4F5TN5E2"},"agent_actions":{"view_html":"https://pith.science/pith/4F5TN5E2AD24QEOPPNGG7ULCGH","download_json":"https://pith.science/pith/4F5TN5E2AD24QEOPPNGG7ULCGH.json","view_paper":"https://pith.science/paper/4F5TN5E2","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2001.01900&json=true","fetch_graph":"https://pith.science/api/pith-number/4F5TN5E2AD24QEOPPNGG7ULCGH/graph.json","fetch_events":"https://pith.science/api/pith-number/4F5TN5E2AD24QEOPPNGG7ULCGH/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/4F5TN5E2AD24QEOPPNGG7ULCGH/action/timestamp_anchor","attest_storage":"https://pith.science/pith/4F5TN5E2AD24QEOPPNGG7ULCGH/action/storage_attestation","attest_author":"https://pith.science/pith/4F5TN5E2AD24QEOPPNGG7ULCGH/action/author_attestation","sign_citation":"https://pith.science/pith/4F5TN5E2AD24QEOPPNGG7ULCGH/action/citation_signature","submit_replication":"https://pith.science/pith/4F5TN5E2AD24QEOPPNGG7ULCGH/action/replication_record"}},"created_at":"2026-07-05T01:16:11.766017+00:00","updated_at":"2026-07-05T01:16:11.766017+00:00"}