{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:WKAA4YFOQ5MIEUVG667M7O7CZD","short_pith_number":"pith:WKAA4YFO","schema_version":"1.0","canonical_sha256":"b2800e60ae87588252a6f7becfbbe2c8da7b6283a0f5bb08efcc224e54d46437","source":{"kind":"arxiv","id":"2206.04777","version":3},"attestation_state":"computed","paper":{"title":"Trimmed Maximum Likelihood Estimation for Robust Learning in Generalized Linear Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Abhimanyu Das, Pranjal Awasthi, Rajat Sen, Weihao Kong","submitted_at":"2022-06-09T21:56:44Z","abstract_excerpt":"We study the problem of learning generalized linear models under adversarial corruptions. We analyze a classical heuristic called the iterative trimmed maximum likelihood estimator which is known to be effective against label corruptions in practice. Under label corruptions, we prove that this simple estimator achieves minimax near-optimal risk on a wide range of generalized linear models, including Gaussian regression, Poisson regression and Binomial regression. Finally, we extend the estimator to the more challenging setting of label and covariate corruptions and demonstrate its robustness a"},"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":"2206.04777","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2022-06-09T21:56:44Z","cross_cats_sorted":["stat.ML"],"title_canon_sha256":"ca073267f45d2ffa441abfb23c2e94cf0b97f7fe20d7a56e0fd0baa1c84da0e9","abstract_canon_sha256":"b0e8a358d712d4bdf2a4a18204fae95987b0d77e37455e8648f7eb882d6346a2"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T05:09:19.166692Z","signature_b64":"2FnjUZnHQi9R5GMaAQEtJJVAeDmdxbwaZQbRzz4GkLQ4+k/0VidQCDv/MvO/xvy+vIthfzkMPrc+ADCXohvnBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"b2800e60ae87588252a6f7becfbbe2c8da7b6283a0f5bb08efcc224e54d46437","last_reissued_at":"2026-07-05T05:09:19.166261Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T05:09:19.166261Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Trimmed Maximum Likelihood Estimation for Robust Learning in Generalized Linear Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Abhimanyu Das, Pranjal Awasthi, Rajat Sen, Weihao Kong","submitted_at":"2022-06-09T21:56:44Z","abstract_excerpt":"We study the problem of learning generalized linear models under adversarial corruptions. We analyze a classical heuristic called the iterative trimmed maximum likelihood estimator which is known to be effective against label corruptions in practice. Under label corruptions, we prove that this simple estimator achieves minimax near-optimal risk on a wide range of generalized linear models, including Gaussian regression, Poisson regression and Binomial regression. Finally, we extend the estimator to the more challenging setting of label and covariate corruptions and demonstrate its robustness a"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2206.04777","kind":"arxiv","version":3},"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/2206.04777/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":"2206.04777","created_at":"2026-07-05T05:09:19.166328+00:00"},{"alias_kind":"arxiv_version","alias_value":"2206.04777v3","created_at":"2026-07-05T05:09:19.166328+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2206.04777","created_at":"2026-07-05T05:09:19.166328+00:00"},{"alias_kind":"pith_short_12","alias_value":"WKAA4YFOQ5MI","created_at":"2026-07-05T05:09:19.166328+00:00"},{"alias_kind":"pith_short_16","alias_value":"WKAA4YFOQ5MIEUVG","created_at":"2026-07-05T05:09:19.166328+00:00"},{"alias_kind":"pith_short_8","alias_value":"WKAA4YFO","created_at":"2026-07-05T05:09:19.166328+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2505.15694","citing_title":"A Unified Theoretical Analysis of Private and Robust Offline Alignment: from RLHF to DPO","ref_index":2021,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/WKAA4YFOQ5MIEUVG667M7O7CZD","json":"https://pith.science/pith/WKAA4YFOQ5MIEUVG667M7O7CZD.json","graph_json":"https://pith.science/api/pith-number/WKAA4YFOQ5MIEUVG667M7O7CZD/graph.json","events_json":"https://pith.science/api/pith-number/WKAA4YFOQ5MIEUVG667M7O7CZD/events.json","paper":"https://pith.science/paper/WKAA4YFO"},"agent_actions":{"view_html":"https://pith.science/pith/WKAA4YFOQ5MIEUVG667M7O7CZD","download_json":"https://pith.science/pith/WKAA4YFOQ5MIEUVG667M7O7CZD.json","view_paper":"https://pith.science/paper/WKAA4YFO","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2206.04777&json=true","fetch_graph":"https://pith.science/api/pith-number/WKAA4YFOQ5MIEUVG667M7O7CZD/graph.json","fetch_events":"https://pith.science/api/pith-number/WKAA4YFOQ5MIEUVG667M7O7CZD/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/WKAA4YFOQ5MIEUVG667M7O7CZD/action/timestamp_anchor","attest_storage":"https://pith.science/pith/WKAA4YFOQ5MIEUVG667M7O7CZD/action/storage_attestation","attest_author":"https://pith.science/pith/WKAA4YFOQ5MIEUVG667M7O7CZD/action/author_attestation","sign_citation":"https://pith.science/pith/WKAA4YFOQ5MIEUVG667M7O7CZD/action/citation_signature","submit_replication":"https://pith.science/pith/WKAA4YFOQ5MIEUVG667M7O7CZD/action/replication_record"}},"created_at":"2026-07-05T05:09:19.166328+00:00","updated_at":"2026-07-05T05:09:19.166328+00:00"}