{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:QADOI6K5SBHT5E4EXZ6BUDQO7V","short_pith_number":"pith:QADOI6K5","schema_version":"1.0","canonical_sha256":"8006e4795d904f3e9384be7c1a0e0efd7ef0753eb8b550c215f87c12ba7fa12e","source":{"kind":"arxiv","id":"2311.02891","version":1},"attestation_state":"computed","paper":{"title":"AdaFlood: Adaptive Flood Regularization","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Danica J. Sutherland, Frederick Tung, Gabriel L. Oliveira, Mohamad Osama Ahmed, Wonho Bae, Yi Ren","submitted_at":"2023-11-06T05:50:10Z","abstract_excerpt":"Although neural networks are conventionally optimized towards zero training loss, it has been recently learned that targeting a non-zero training loss threshold, referred to as a flood level, often enables better test time generalization. Current approaches, however, apply the same constant flood level to all training samples, which inherently assumes all the samples have the same difficulty. We present AdaFlood, a novel flood regularization method that adapts the flood level of each training sample according to the difficulty of the sample. Intuitively, since training samples are not equal in"},"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":"2311.02891","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2023-11-06T05:50:10Z","cross_cats_sorted":[],"title_canon_sha256":"8844a2e6887bf753b5443d3bdd033faa5e71f7d0305f1ae0867817856582c6e6","abstract_canon_sha256":"0d5e860962b25561cd594ec6c74edb1e4a9c05774e17e71f203a97fe5ed55da7"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:09:19.262812Z","signature_b64":"TYUsSE05lyRveamyhQSCCE3gz38f6NROkyI4fN2EmWSBOi2bTVQyzN6VH6MFvgQfl4Hmcq09sCPdd0/USDaFCA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"8006e4795d904f3e9384be7c1a0e0efd7ef0753eb8b550c215f87c12ba7fa12e","last_reissued_at":"2026-07-05T07:09:19.262282Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:09:19.262282Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"AdaFlood: Adaptive Flood Regularization","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Danica J. Sutherland, Frederick Tung, Gabriel L. Oliveira, Mohamad Osama Ahmed, Wonho Bae, Yi Ren","submitted_at":"2023-11-06T05:50:10Z","abstract_excerpt":"Although neural networks are conventionally optimized towards zero training loss, it has been recently learned that targeting a non-zero training loss threshold, referred to as a flood level, often enables better test time generalization. Current approaches, however, apply the same constant flood level to all training samples, which inherently assumes all the samples have the same difficulty. We present AdaFlood, a novel flood regularization method that adapts the flood level of each training sample according to the difficulty of the sample. Intuitively, since training samples are not equal in"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2311.02891","kind":"arxiv","version":1},"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/2311.02891/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":"2311.02891","created_at":"2026-07-05T07:09:19.262359+00:00"},{"alias_kind":"arxiv_version","alias_value":"2311.02891v1","created_at":"2026-07-05T07:09:19.262359+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2311.02891","created_at":"2026-07-05T07:09:19.262359+00:00"},{"alias_kind":"pith_short_12","alias_value":"QADOI6K5SBHT","created_at":"2026-07-05T07:09:19.262359+00:00"},{"alias_kind":"pith_short_16","alias_value":"QADOI6K5SBHT5E4E","created_at":"2026-07-05T07:09:19.262359+00:00"},{"alias_kind":"pith_short_8","alias_value":"QADOI6K5","created_at":"2026-07-05T07:09:19.262359+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/QADOI6K5SBHT5E4EXZ6BUDQO7V","json":"https://pith.science/pith/QADOI6K5SBHT5E4EXZ6BUDQO7V.json","graph_json":"https://pith.science/api/pith-number/QADOI6K5SBHT5E4EXZ6BUDQO7V/graph.json","events_json":"https://pith.science/api/pith-number/QADOI6K5SBHT5E4EXZ6BUDQO7V/events.json","paper":"https://pith.science/paper/QADOI6K5"},"agent_actions":{"view_html":"https://pith.science/pith/QADOI6K5SBHT5E4EXZ6BUDQO7V","download_json":"https://pith.science/pith/QADOI6K5SBHT5E4EXZ6BUDQO7V.json","view_paper":"https://pith.science/paper/QADOI6K5","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2311.02891&json=true","fetch_graph":"https://pith.science/api/pith-number/QADOI6K5SBHT5E4EXZ6BUDQO7V/graph.json","fetch_events":"https://pith.science/api/pith-number/QADOI6K5SBHT5E4EXZ6BUDQO7V/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/QADOI6K5SBHT5E4EXZ6BUDQO7V/action/timestamp_anchor","attest_storage":"https://pith.science/pith/QADOI6K5SBHT5E4EXZ6BUDQO7V/action/storage_attestation","attest_author":"https://pith.science/pith/QADOI6K5SBHT5E4EXZ6BUDQO7V/action/author_attestation","sign_citation":"https://pith.science/pith/QADOI6K5SBHT5E4EXZ6BUDQO7V/action/citation_signature","submit_replication":"https://pith.science/pith/QADOI6K5SBHT5E4EXZ6BUDQO7V/action/replication_record"}},"created_at":"2026-07-05T07:09:19.262359+00:00","updated_at":"2026-07-05T07:09:19.262359+00:00"}