{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:7CSJQHHEQFLJ37FANUKFN3WCRZ","short_pith_number":"pith:7CSJQHHE","schema_version":"1.0","canonical_sha256":"f8a4981ce481569dfca06d1456eec28e64d76fbbaead36c52876ed9477bb1d39","source":{"kind":"arxiv","id":"2607.05903","version":1},"attestation_state":"computed","paper":{"title":"K-ABENA: K-Adaptive Backpropagation with Error-based N-exclusion Algorithm : (Compensated Loss-Based Sample Exclusion with Unbiased Gradient Estimation)","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.CL"],"primary_cat":"cs.LG","authors_text":"Jean-Francois Bonbhel","submitted_at":"2026-07-07T06:58:51Z","abstract_excerpt":"We present K-ABENA (K-Adaptive Backpropagation with Error-based N-exclusion Algorithm), a selective gradient computation framework that reduces per-iteration training cost by excluding a fraction of low-loss (\"minor\") observations from the backward pass. Its canonical form (v3) combines a defensive-mixture sampling design over the minor set with Horvitz-Thompson inverse-probability reweighting, yielding a design-unbiased Horvitz-Thompson gradient estimator (Lemma 2) and whose self-normalized practical variant carries a bias of order O(1/m) with an explicit constant (Lemma 3). We prove an O(1/s"},"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":"2607.05903","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-07-07T06:58:51Z","cross_cats_sorted":["cs.AI","cs.CL"],"title_canon_sha256":"42524294b0415e1b8bcbfce688977474ccc4885fc9e47bc56f483d6c4fb83d8d","abstract_canon_sha256":"2d3763ef5a548ddbbf3ee355038d8b0f959893cda2ee00e937d13b547a49dc6c"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-08T01:18:49.771307Z","signature_b64":"0Q81uFBwtBfKot6acgvgWMiqAZM1LxH8jmCITK/kOicMTUu1e6ek3Ekrhlzbc4UqMRD7W+tCgn7wO2fZ8fTODA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"f8a4981ce481569dfca06d1456eec28e64d76fbbaead36c52876ed9477bb1d39","last_reissued_at":"2026-07-08T01:18:49.770881Z","signature_status":"signed_v1","first_computed_at":"2026-07-08T01:18:49.770881Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"K-ABENA: K-Adaptive Backpropagation with Error-based N-exclusion Algorithm : (Compensated Loss-Based Sample Exclusion with Unbiased Gradient Estimation)","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.CL"],"primary_cat":"cs.LG","authors_text":"Jean-Francois Bonbhel","submitted_at":"2026-07-07T06:58:51Z","abstract_excerpt":"We present K-ABENA (K-Adaptive Backpropagation with Error-based N-exclusion Algorithm), a selective gradient computation framework that reduces per-iteration training cost by excluding a fraction of low-loss (\"minor\") observations from the backward pass. Its canonical form (v3) combines a defensive-mixture sampling design over the minor set with Horvitz-Thompson inverse-probability reweighting, yielding a design-unbiased Horvitz-Thompson gradient estimator (Lemma 2) and whose self-normalized practical variant carries a bias of order O(1/m) with an explicit constant (Lemma 3). We prove an O(1/s"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.05903","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/2607.05903/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":"2607.05903","created_at":"2026-07-08T01:18:49.770947+00:00"},{"alias_kind":"arxiv_version","alias_value":"2607.05903v1","created_at":"2026-07-08T01:18:49.770947+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.05903","created_at":"2026-07-08T01:18:49.770947+00:00"},{"alias_kind":"pith_short_12","alias_value":"7CSJQHHEQFLJ","created_at":"2026-07-08T01:18:49.770947+00:00"},{"alias_kind":"pith_short_16","alias_value":"7CSJQHHEQFLJ37FA","created_at":"2026-07-08T01:18:49.770947+00:00"},{"alias_kind":"pith_short_8","alias_value":"7CSJQHHE","created_at":"2026-07-08T01:18:49.770947+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/7CSJQHHEQFLJ37FANUKFN3WCRZ","json":"https://pith.science/pith/7CSJQHHEQFLJ37FANUKFN3WCRZ.json","graph_json":"https://pith.science/api/pith-number/7CSJQHHEQFLJ37FANUKFN3WCRZ/graph.json","events_json":"https://pith.science/api/pith-number/7CSJQHHEQFLJ37FANUKFN3WCRZ/events.json","paper":"https://pith.science/paper/7CSJQHHE"},"agent_actions":{"view_html":"https://pith.science/pith/7CSJQHHEQFLJ37FANUKFN3WCRZ","download_json":"https://pith.science/pith/7CSJQHHEQFLJ37FANUKFN3WCRZ.json","view_paper":"https://pith.science/paper/7CSJQHHE","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2607.05903&json=true","fetch_graph":"https://pith.science/api/pith-number/7CSJQHHEQFLJ37FANUKFN3WCRZ/graph.json","fetch_events":"https://pith.science/api/pith-number/7CSJQHHEQFLJ37FANUKFN3WCRZ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/7CSJQHHEQFLJ37FANUKFN3WCRZ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/7CSJQHHEQFLJ37FANUKFN3WCRZ/action/storage_attestation","attest_author":"https://pith.science/pith/7CSJQHHEQFLJ37FANUKFN3WCRZ/action/author_attestation","sign_citation":"https://pith.science/pith/7CSJQHHEQFLJ37FANUKFN3WCRZ/action/citation_signature","submit_replication":"https://pith.science/pith/7CSJQHHEQFLJ37FANUKFN3WCRZ/action/replication_record"}},"created_at":"2026-07-08T01:18:49.770947+00:00","updated_at":"2026-07-08T01:18:49.770947+00:00"}