{"id":"https://openalex.org/W7161272297","doi":"https://doi.org/10.48550/arxiv.2605.15134","title":"Training ML Models with Predictable Failures","display_name":"Training ML Models with Predictable Failures","publication_year":2026,"publication_date":"2026-05-14","ids":{"openalex":"https://openalex.org/W7161272297","doi":"https://doi.org/10.48550/arxiv.2605.15134"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2605.15134","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.15134","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"article"},"type":"preprint","indexed_in":["datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://doi.org/10.48550/arxiv.2605.15134","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5038628759","display_name":"Will Schwarzer","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Schwarzer, Will","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5136251479","display_name":"Scott Niekum","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Niekum, Scott","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":2,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":null,"last_page":null},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T12127","display_name":"Software System Performance and Reliability","score":0.5309000015258789,"subfield":{"id":"https://openalex.org/subfields/1705","display_name":"Computer Networks and Communications"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},"topics":[{"id":"https://openalex.org/T12127","display_name":"Software System Performance and Reliability","score":0.5309000015258789,"subfield":{"id":"https://openalex.org/subfields/1705","display_name":"Computer Networks and Communications"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T11689","display_name":"Adversarial Robustness in Machine Learning","score":0.06319999694824219,"subfield":{"id":"https://openalex.org/subfields/1702","display_name":"Artificial Intelligence"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T14347","display_name":"Big Data and Digital Economy","score":0.020899999886751175,"subfield":{"id":"https://openalex.org/subfields/1710","display_name":"Information Systems"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/software-deployment","display_name":"Software deployment","score":0.7760999798774719},{"id":"https://openalex.org/keywords/offset","display_name":"Offset (computer science)","score":0.6417999863624573},{"id":"https://openalex.org/keywords/set","display_name":"Set (abstract data type)","score":0.5533000230789185},{"id":"https://openalex.org/keywords/scale","display_name":"Scale (ratio)","score":0.4514000117778778},{"id":"https://openalex.org/keywords/key","display_name":"Key (lock)","score":0.396699994802475},{"id":"https://openalex.org/keywords/failure-mode-and-effects-analysis","display_name":"Failure mode and effects analysis","score":0.38760000467300415},{"id":"https://openalex.org/keywords/mode","display_name":"Mode (computer interface)","score":0.32670000195503235},{"id":"https://openalex.org/keywords/training","display_name":"Training (meteorology)","score":0.3160000145435333}],"concepts":[{"id":"https://openalex.org/C105339364","wikidata":"https://www.wikidata.org/wiki/Q2297740","display_name":"Software deployment","level":2,"score":0.7760999798774719},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6747000217437744},{"id":"https://openalex.org/C175291020","wikidata":"https://www.wikidata.org/wiki/Q1156822","display_name":"Offset (computer science)","level":2,"score":0.6417999863624573},{"id":"https://openalex.org/C177264268","wikidata":"https://www.wikidata.org/wiki/Q1514741","display_name":"Set (abstract data type)","level":2,"score":0.5533000230789185},{"id":"https://openalex.org/C2778755073","wikidata":"https://www.wikidata.org/wiki/Q10858537","display_name":"Scale (ratio)","level":2,"score":0.4514000117778778},{"id":"https://openalex.org/C200601418","wikidata":"https://www.wikidata.org/wiki/Q2193887","display_name":"Reliability engineering","level":1,"score":0.40700000524520874},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.396699994802475},{"id":"https://openalex.org/C66283442","wikidata":"https://www.wikidata.org/wiki/Q1389268","display_name":"Failure mode and effects analysis","level":2,"score":0.38760000467300415},{"id":"https://openalex.org/C48677424","wikidata":"https://www.wikidata.org/wiki/Q6888088","display_name":"Mode (computer interface)","level":2,"score":0.32670000195503235},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.31690001487731934},{"id":"https://openalex.org/C2777211547","wikidata":"https://www.wikidata.org/wiki/Q17141490","display_name":"Training (meteorology)","level":2,"score":0.3160000145435333},{"id":"https://openalex.org/C51632099","wikidata":"https://www.wikidata.org/wiki/Q3985153","display_name":"Training set","level":2,"score":0.30660000443458557},{"id":"https://openalex.org/C2780009758","wikidata":"https://www.wikidata.org/wiki/Q6804172","display_name":"Measure (data warehouse)","level":2,"score":0.29280000925064087},{"id":"https://openalex.org/C174303752","wikidata":"https://www.wikidata.org/wiki/Q3624438","display_name":"Forecast error","level":2,"score":0.2865999937057495},{"id":"https://openalex.org/C124101348","wikidata":"https://www.wikidata.org/wiki/Q172491","display_name":"Data mining","level":1,"score":0.2678999900817871},{"id":"https://openalex.org/C103088060","wikidata":"https://www.wikidata.org/wiki/Q1062839","display_name":"Error detection and correction","level":2,"score":0.2614000141620636},{"id":"https://openalex.org/C63479239","wikidata":"https://www.wikidata.org/wiki/Q7353546","display_name":"Robustness (evolution)","level":3,"score":0.2606000006198883},{"id":"https://openalex.org/C124681953","wikidata":"https://www.wikidata.org/wiki/Q339062","display_name":"Decomposition","level":2,"score":0.25690001249313354},{"id":"https://openalex.org/C2779662365","wikidata":"https://www.wikidata.org/wiki/Q5416694","display_name":"Event (particle physics)","level":2,"score":0.25619998574256897},{"id":"https://openalex.org/C167085575","wikidata":"https://www.wikidata.org/wiki/Q6803654","display_name":"Mean squared prediction error","level":2,"score":0.2524999976158142}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2605.15134","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.15134","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":null,"is_accepted":false,"is_published":null,"raw_source_name":null,"raw_type":"article"}],"best_oa_location":{"id":"doi:10.48550/arxiv.2605.15134","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.15134","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"article"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Estimating":[0],"how":[1],"often":[2],"an":[3,47,132],"ML":[4],"model":[5],"will":[6],"fail":[7],"at":[8,107],"deployment":[9,99,108],"scale":[10],"is":[11,22,80,86],"central":[12],"to":[13,26,50,105,149],"pre-deployment":[14],"safety":[15,147],"assessment,":[16],"but":[17],"a":[18,57,70,93,112,127],"feasible":[19],"evaluation":[20,48,90],"set":[21,49,91,100],"rarely":[23],"large":[24],"enough":[25],"observe":[27],"the":[28,41,76,81,89,98,103,115],"failures":[29],"that":[30,67,97,118,150],"matter.":[31],"Jones":[32],"et":[33],"al.":[34],"(2025)":[35],"address":[36],"this":[37,61,120],"by":[38],"extrapolating":[39],"from":[40],"largest":[42],"k":[43],"failure":[44,53,121],"scores":[45],"in":[46,75],"predict":[51],"deployment-scale":[52],"rates.":[54],"We":[55,110],"give":[56],"finite-k":[58],"decomposition":[59],"of":[60,151],"estimator's":[62],"forecast":[63,104,139],"error":[64,140],"and":[65,131,145],"show":[66],"it":[68],"has":[69],"built-in":[71],"bias":[72,85],"toward":[73],"over-prediction":[74],"typical":[77],"case,":[78],"which":[79],"safety-favorable":[82],"direction.":[83],"This":[84],"offset":[87],"when":[88],"misses":[92],"rare":[94],"high-failure":[95],"mode":[96],"contains,":[101],"leaving":[102],"under-predict":[106],"scale.":[109],"propose":[111],"fine-tuning":[113,135],"objective,":[114],"forecastability":[116],"loss,":[117],"addresses":[119],"mode.":[122],"In":[123],"two":[124],"proof-of-concept":[125],"experiments,":[126],"language-model":[128],"password":[129],"game":[130],"RL":[133],"gridworld,":[134],"substantially":[136],"reduces":[137],"held-out":[138],"while":[141],"preserving":[142],"primary-task":[143],"capability":[144],"achieving":[146],"similar":[148],"supervised":[152],"baselines.":[153]},"counts_by_year":[],"updated_date":"2026-06-11T09:08:48.828518","created_date":"2026-05-16T00:00:00"}