{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2020:3OE4WDTTAFJYYH7Q2IM7GIHUYW","short_pith_number":"pith:3OE4WDTT","schema_version":"1.0","canonical_sha256":"db89cb0e7301538c1ff0d219f320f4c5b4540cdd611d00e7087e7f46f59584d4","source":{"kind":"arxiv","id":"2009.06111","version":2},"attestation_state":"computed","paper":{"title":"Machine Learning's Dropout Training is Distributionally Robust Optimal","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"stat.ML","authors_text":"Jose Blanchet, Jose Luis Montiel Olea, Viet Anh Nguyen, Xuhui Zhang, Yang Kang","submitted_at":"2020-09-13T23:13:28Z","abstract_excerpt":"This paper shows that dropout training in Generalized Linear Models is the minimax solution of a two-player, zero-sum game where an adversarial nature corrupts a statistician's covariates using a multiplicative nonparametric errors-in-variables model. In this game, nature's least favorable distribution is dropout noise, where nature independently deletes entries of the covariate vector with some fixed probability $\\delta$. This result implies that dropout training indeed provides out-of-sample expected loss guarantees for distributions that arise from multiplicative perturbations of in-sample "},"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":"2009.06111","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"stat.ML","submitted_at":"2020-09-13T23:13:28Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"52d0672e9e362547b71f831c5441e85a825508806062c60f5054c79fd0201fd4","abstract_canon_sha256":"85a87eba62051710865870e0a36adbf0b3b04d2fffc5a42ff7e2be1959f8e916"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T02:32:00.856217Z","signature_b64":"34AqH3LV/DzRG9jy9iwZFTquXYTmEOydYwS8tWB6IvXxYMf5/DjhZTcTPfNJYf4pdgKZh8IfozL+n4sCCQ5pCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"db89cb0e7301538c1ff0d219f320f4c5b4540cdd611d00e7087e7f46f59584d4","last_reissued_at":"2026-07-05T02:32:00.855781Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T02:32:00.855781Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Machine Learning's Dropout Training is Distributionally Robust Optimal","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"stat.ML","authors_text":"Jose Blanchet, Jose Luis Montiel Olea, Viet Anh Nguyen, Xuhui Zhang, Yang Kang","submitted_at":"2020-09-13T23:13:28Z","abstract_excerpt":"This paper shows that dropout training in Generalized Linear Models is the minimax solution of a two-player, zero-sum game where an adversarial nature corrupts a statistician's covariates using a multiplicative nonparametric errors-in-variables model. In this game, nature's least favorable distribution is dropout noise, where nature independently deletes entries of the covariate vector with some fixed probability $\\delta$. This result implies that dropout training indeed provides out-of-sample expected loss guarantees for distributions that arise from multiplicative perturbations of in-sample "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2009.06111","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/2009.06111/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":"2009.06111","created_at":"2026-07-05T02:32:00.855847+00:00"},{"alias_kind":"arxiv_version","alias_value":"2009.06111v2","created_at":"2026-07-05T02:32:00.855847+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2009.06111","created_at":"2026-07-05T02:32:00.855847+00:00"},{"alias_kind":"pith_short_12","alias_value":"3OE4WDTTAFJY","created_at":"2026-07-05T02:32:00.855847+00:00"},{"alias_kind":"pith_short_16","alias_value":"3OE4WDTTAFJYYH7Q","created_at":"2026-07-05T02:32:00.855847+00:00"},{"alias_kind":"pith_short_8","alias_value":"3OE4WDTT","created_at":"2026-07-05T02:32:00.855847+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/3OE4WDTTAFJYYH7Q2IM7GIHUYW","json":"https://pith.science/pith/3OE4WDTTAFJYYH7Q2IM7GIHUYW.json","graph_json":"https://pith.science/api/pith-number/3OE4WDTTAFJYYH7Q2IM7GIHUYW/graph.json","events_json":"https://pith.science/api/pith-number/3OE4WDTTAFJYYH7Q2IM7GIHUYW/events.json","paper":"https://pith.science/paper/3OE4WDTT"},"agent_actions":{"view_html":"https://pith.science/pith/3OE4WDTTAFJYYH7Q2IM7GIHUYW","download_json":"https://pith.science/pith/3OE4WDTTAFJYYH7Q2IM7GIHUYW.json","view_paper":"https://pith.science/paper/3OE4WDTT","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2009.06111&json=true","fetch_graph":"https://pith.science/api/pith-number/3OE4WDTTAFJYYH7Q2IM7GIHUYW/graph.json","fetch_events":"https://pith.science/api/pith-number/3OE4WDTTAFJYYH7Q2IM7GIHUYW/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/3OE4WDTTAFJYYH7Q2IM7GIHUYW/action/timestamp_anchor","attest_storage":"https://pith.science/pith/3OE4WDTTAFJYYH7Q2IM7GIHUYW/action/storage_attestation","attest_author":"https://pith.science/pith/3OE4WDTTAFJYYH7Q2IM7GIHUYW/action/author_attestation","sign_citation":"https://pith.science/pith/3OE4WDTTAFJYYH7Q2IM7GIHUYW/action/citation_signature","submit_replication":"https://pith.science/pith/3OE4WDTTAFJYYH7Q2IM7GIHUYW/action/replication_record"}},"created_at":"2026-07-05T02:32:00.855847+00:00","updated_at":"2026-07-05T02:32:00.855847+00:00"}