{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2020:LQVE2UF464FJ6KU77YTFENBGX7","short_pith_number":"pith:LQVE2UF4","schema_version":"1.0","canonical_sha256":"5c2a4d50bcf70a9f2a9ffe26523426bff6614489a57b9f35d0df36652419a8f8","source":{"kind":"arxiv","id":"2002.08709","version":2},"attestation_state":"computed","paper":{"title":"Do We Need Zero Training Loss After Achieving Zero Training Error?","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Gang Niu, Ikko Yamane, Masashi Sugiyama, Takashi Ishida, Tomoya Sakai","submitted_at":"2020-02-20T12:50:49Z","abstract_excerpt":"Overparameterized deep networks have the capacity to memorize training data with zero \\emph{training error}. Even after memorization, the \\emph{training loss} continues to approach zero, making the model overconfident and the test performance degraded. Since existing regularizers do not directly aim to avoid zero training loss, it is hard to tune their hyperparameters in order to maintain a fixed/preset level of training loss. We propose a direct solution called \\emph{flooding} that intentionally prevents further reduction of the training loss when it reaches a reasonably small value, which we"},"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":"2002.08709","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2020-02-20T12:50:49Z","cross_cats_sorted":["stat.ML"],"title_canon_sha256":"9aff662a85e8b0fb2b930fa62d0cd2efb3c2cc3d14dedc45bf94163a566d87a3","abstract_canon_sha256":"c2ec1f9778983f6460df86d26c3ad1d6ce96dc430db668d727a29448186dd396"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T02:28:04.446743Z","signature_b64":"7Vraktmjo6HzyxHlyh1YMSicBcns6zAG2Ta9WVw8/gUFgX1tbdqz31E+6dDg+cYis7xl0hgSOL7wBiu0dgwxBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"5c2a4d50bcf70a9f2a9ffe26523426bff6614489a57b9f35d0df36652419a8f8","last_reissued_at":"2026-07-05T02:28:04.446189Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T02:28:04.446189Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Do We Need Zero Training Loss After Achieving Zero Training Error?","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Gang Niu, Ikko Yamane, Masashi Sugiyama, Takashi Ishida, Tomoya Sakai","submitted_at":"2020-02-20T12:50:49Z","abstract_excerpt":"Overparameterized deep networks have the capacity to memorize training data with zero \\emph{training error}. Even after memorization, the \\emph{training loss} continues to approach zero, making the model overconfident and the test performance degraded. Since existing regularizers do not directly aim to avoid zero training loss, it is hard to tune their hyperparameters in order to maintain a fixed/preset level of training loss. We propose a direct solution called \\emph{flooding} that intentionally prevents further reduction of the training loss when it reaches a reasonably small value, which we"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2002.08709","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/2002.08709/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":"2002.08709","created_at":"2026-07-05T02:28:04.446282+00:00"},{"alias_kind":"arxiv_version","alias_value":"2002.08709v2","created_at":"2026-07-05T02:28:04.446282+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2002.08709","created_at":"2026-07-05T02:28:04.446282+00:00"},{"alias_kind":"pith_short_12","alias_value":"LQVE2UF464FJ","created_at":"2026-07-05T02:28:04.446282+00:00"},{"alias_kind":"pith_short_16","alias_value":"LQVE2UF464FJ6KU7","created_at":"2026-07-05T02:28:04.446282+00:00"},{"alias_kind":"pith_short_8","alias_value":"LQVE2UF4","created_at":"2026-07-05T02:28:04.446282+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/LQVE2UF464FJ6KU77YTFENBGX7","json":"https://pith.science/pith/LQVE2UF464FJ6KU77YTFENBGX7.json","graph_json":"https://pith.science/api/pith-number/LQVE2UF464FJ6KU77YTFENBGX7/graph.json","events_json":"https://pith.science/api/pith-number/LQVE2UF464FJ6KU77YTFENBGX7/events.json","paper":"https://pith.science/paper/LQVE2UF4"},"agent_actions":{"view_html":"https://pith.science/pith/LQVE2UF464FJ6KU77YTFENBGX7","download_json":"https://pith.science/pith/LQVE2UF464FJ6KU77YTFENBGX7.json","view_paper":"https://pith.science/paper/LQVE2UF4","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2002.08709&json=true","fetch_graph":"https://pith.science/api/pith-number/LQVE2UF464FJ6KU77YTFENBGX7/graph.json","fetch_events":"https://pith.science/api/pith-number/LQVE2UF464FJ6KU77YTFENBGX7/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/LQVE2UF464FJ6KU77YTFENBGX7/action/timestamp_anchor","attest_storage":"https://pith.science/pith/LQVE2UF464FJ6KU77YTFENBGX7/action/storage_attestation","attest_author":"https://pith.science/pith/LQVE2UF464FJ6KU77YTFENBGX7/action/author_attestation","sign_citation":"https://pith.science/pith/LQVE2UF464FJ6KU77YTFENBGX7/action/citation_signature","submit_replication":"https://pith.science/pith/LQVE2UF464FJ6KU77YTFENBGX7/action/replication_record"}},"created_at":"2026-07-05T02:28:04.446282+00:00","updated_at":"2026-07-05T02:28:04.446282+00:00"}