{"id":"https://openalex.org/W7214728114","doi":"https://doi.org/10.48550/arxiv.2609.33377","title":"Optimal Transport Dropout for Structured Predictive Uncertainty","display_name":"Optimal Transport Dropout for Structured Predictive Uncertainty","publication_year":2026,"publication_date":"2026-09-27","ids":{"openalex":"https://openalex.org/W7214728114","doi":"https://doi.org/10.48550/arxiv.2609.33377"},"language":"en","primary_location":{"id":"pmh:oai:arXiv.org:2609.33377","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2609.33377","pdf_url":"https://arxiv.org/pdf/2609.33377","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"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":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"volume":null,"type":"preprint","is_traditional":true,"indexed_in":["arxiv","datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2609.33377","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5106356256","display_name":"Giacomo Lorenzon","orcid":"https://orcid.org/0009-0008-5188-3719","observed_orcids":["https://orcid.org/0009-0008-5188-3719"]},"institutions":[{"id":"https://openalex.org/I93860229","display_name":"Politecnico di Milano","ror":"https://ror.org/01nffqt88","country_code":"IT","type":"education","lineage":["https://openalex.org/I93860229"]}],"countries":["IT"],"is_corresponding":false,"raw_author_name":"Lorenzon, Giacomo","raw_affiliation_strings":["MOX-Department of Mathematics , Politecnico di Milano , Italy"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"MOX-Department of Mathematics , Politecnico di Milano , Italy","institution_ids":["https://openalex.org/I93860229"]}]},{"author_position":"last","author":{"id":null,"display_name":"Regazzoni, Francesco","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I93860229","display_name":"Politecnico di Milano","ror":"https://ror.org/01nffqt88","country_code":"IT","type":"education","lineage":["https://openalex.org/I93860229"]}],"countries":["IT"],"is_corresponding":false,"raw_author_name":"Regazzoni, Francesco","raw_affiliation_strings":["MOX-Department of Mathematics , Politecnico di Milano , Italy"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"MOX-Department of Mathematics , Politecnico di Milano , Italy","institution_ids":["https://openalex.org/I93860229"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I93860229"],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":true,"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/T12814","display_name":"Gaussian Processes and Bayesian Inference","score":0.7271000146865845,"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"}},"topics":[{"id":"https://openalex.org/T12814","display_name":"Gaussian Processes and Bayesian Inference","score":0.7271000146865845,"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/T11689","display_name":"Adversarial Robustness in Machine Learning","score":0.20229999721050262,"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/T11206","display_name":"Model Reduction and Neural Networks","score":0.022199999541044235,"subfield":{"id":"https://openalex.org/subfields/3109","display_name":"Statistical and Nonlinear Physics"},"field":{"id":"https://openalex.org/fields/31","display_name":"Physics and Astronomy"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/monte-carlo-dropout","display_name":"Monte Carlo dropout","score":0.890999972820282},{"id":"https://openalex.org/keywords/forecast-uncertainty","display_name":"forecast uncertainty","score":0.8939999938011169},{"id":"https://openalex.org/keywords/neural-operators","display_name":"neural operators","score":0.7139999866485596},{"id":"https://openalex.org/keywords/probabilistic-neural-network","display_name":"probabilistic neural network","score":0.5899999737739563},{"id":"https://openalex.org/keywords/energy-score","display_name":"energy score","score":0.6830000281333923},{"id":"https://openalex.org/keywords/optimal-transport","display_name":"optimal transport","score":0.4490000009536743},{"id":"https://openalex.org/keywords/model-misspecification","display_name":"model misspecification","score":0.7749999761581421},{"id":"https://openalex.org/keywords/regression-analysis","display_name":"regression analysis","score":0.3070000112056732},{"id":"https://openalex.org/keywords/uncertainty-calibration","display_name":"uncertainty calibration","score":0.6380000114440918}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.5437999963760376},{"id":"https://openalex.org/C19499675","wikidata":"https://www.wikidata.org/wiki/Q232207","display_name":"Monte Carlo method","level":2,"score":0.5389000177383423},{"id":"https://openalex.org/C917703","wikidata":"https://www.wikidata.org/wiki/Q7239668","display_name":"Predictive inference","level":5,"score":0.5182999968528748},{"id":"https://openalex.org/C51167844","wikidata":"https://www.wikidata.org/wiki/Q4422623","display_name":"Latent variable","level":2,"score":0.4916999936103821},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.4772999882698059},{"id":"https://openalex.org/C75553542","wikidata":"https://www.wikidata.org/wiki/Q178161","display_name":"A priori and a posteriori","level":2,"score":0.4327000081539154},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.42250001430511475},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.4146000146865845},{"id":"https://openalex.org/C160234255","wikidata":"https://www.wikidata.org/wiki/Q812535","display_name":"Bayesian inference","level":3,"score":0.40230000019073486},{"id":"https://openalex.org/C107673813","wikidata":"https://www.wikidata.org/wiki/Q812534","display_name":"Bayesian probability","level":2,"score":0.4009999930858612},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.4007999897003174},{"id":"https://openalex.org/C57830394","wikidata":"https://www.wikidata.org/wiki/Q278079","display_name":"Posterior probability","level":3,"score":0.39719998836517334},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.3959999978542328},{"id":"https://openalex.org/C32230216","wikidata":"https://www.wikidata.org/wiki/Q7882499","display_name":"Uncertainty quantification","level":2,"score":0.39079999923706055},{"id":"https://openalex.org/C2776145597","wikidata":"https://www.wikidata.org/wiki/Q25339462","display_name":"Dropout (neural networks)","level":2,"score":0.3709000051021576},{"id":"https://openalex.org/C49937458","wikidata":"https://www.wikidata.org/wiki/Q2599292","display_name":"Probabilistic logic","level":2,"score":0.3366999924182892},{"id":"https://openalex.org/C172205157","wikidata":"https://www.wikidata.org/wiki/Q1782962","display_name":"Model predictive control","level":3,"score":0.31299999356269836},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.3102000057697296},{"id":"https://openalex.org/C52421305","wikidata":"https://www.wikidata.org/wiki/Q1151499","display_name":"Particle filter","level":3,"score":0.3061999976634979},{"id":"https://openalex.org/C61797465","wikidata":"https://www.wikidata.org/wiki/Q1188986","display_name":"Term (time)","level":2,"score":0.303600013256073},{"id":"https://openalex.org/C126255220","wikidata":"https://www.wikidata.org/wiki/Q141495","display_name":"Mathematical optimization","level":1,"score":0.29750001430511475},{"id":"https://openalex.org/C2776401178","wikidata":"https://www.wikidata.org/wiki/Q12050496","display_name":"Feature (linguistics)","level":2,"score":0.29350000619888306},{"id":"https://openalex.org/C61326573","wikidata":"https://www.wikidata.org/wiki/Q1496376","display_name":"Gaussian process","level":3,"score":0.29319998621940613},{"id":"https://openalex.org/C114289077","wikidata":"https://www.wikidata.org/wiki/Q3284399","display_name":"Statistical model","level":2,"score":0.28189998865127563},{"id":"https://openalex.org/C177769412","wikidata":"https://www.wikidata.org/wiki/Q278090","display_name":"Prior probability","level":3,"score":0.2815999984741211},{"id":"https://openalex.org/C177918212","wikidata":"https://www.wikidata.org/wiki/Q803623","display_name":"Perturbation (astronomy)","level":2,"score":0.28119999170303345},{"id":"https://openalex.org/C163716315","wikidata":"https://www.wikidata.org/wiki/Q901177","display_name":"Gaussian","level":2,"score":0.27469998598098755},{"id":"https://openalex.org/C122123141","wikidata":"https://www.wikidata.org/wiki/Q176623","display_name":"Random variable","level":2,"score":0.27239999175071716},{"id":"https://openalex.org/C51955184","wikidata":"https://www.wikidata.org/wiki/Q1545585","display_name":"Stochastic differential equation","level":2,"score":0.26969999074935913},{"id":"https://openalex.org/C2780009758","wikidata":"https://www.wikidata.org/wiki/Q6804172","display_name":"Measure (data warehouse)","level":2,"score":0.26910001039505005},{"id":"https://openalex.org/C153083717","wikidata":"https://www.wikidata.org/wiki/Q6535263","display_name":"Leverage (statistics)","level":2,"score":0.26489999890327454},{"id":"https://openalex.org/C88871306","wikidata":"https://www.wikidata.org/wiki/Q7208287","display_name":"Point process","level":2,"score":0.26089999079704285},{"id":"https://openalex.org/C79772020","wikidata":"https://www.wikidata.org/wiki/Q5159264","display_name":"Conditional independence","level":2,"score":0.25999999046325684},{"id":"https://openalex.org/C2779363554","wikidata":"https://www.wikidata.org/wiki/Q5420835","display_name":"Expectation propagation","level":4,"score":0.2590000033378601},{"id":"https://openalex.org/C163175372","wikidata":"https://www.wikidata.org/wiki/Q3339222","display_name":"Linear model","level":2,"score":0.25619998574256897},{"id":"https://openalex.org/C83546350","wikidata":"https://www.wikidata.org/wiki/Q1139051","display_name":"Regression","level":2,"score":0.25380000472068787},{"id":"https://openalex.org/C83247935","wikidata":"https://www.wikidata.org/wiki/Q7234227","display_name":"Posterior predictive distribution","level":5,"score":0.25029999017715454}],"mesh":[],"locations_count":2,"locations":[{"id":"pmh:oai:arXiv.org:2609.33377","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2609.33377","pdf_url":"https://arxiv.org/pdf/2609.33377","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"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":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},{"id":"doi:10.48550/arxiv.2609.33377","doi":"https://doi.org/10.48550/arxiv.2609.33377","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2609.33377","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"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":"Preprint"}],"best_oa_location":{"id":"pmh:oai:arXiv.org:2609.33377","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2609.33377","pdf_url":"https://arxiv.org/pdf/2609.33377","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"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":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"sustainable_development_goals":[],"study_designs":[],"awards":[],"funders":[{"id":"https://openalex.org/F4320311030","display_name":"Istituto Nazionale di Alta Matematica \"Francesco Severi\"","ror":"https://ror.org/01vx64p53"},{"id":"https://openalex.org/F4320334079","display_name":"Gruppo Nazionale per il Calcolo Scientifico","ror":"https://ror.org/02rnys730"}],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W7214728114.pdf","grobid_xml":"https://content.openalex.org/works/W7214728114.grobid-xml"},"referenced_works_count":8,"referenced_works":["https://openalex.org/W4416549855","https://openalex.org/W7083685169","https://openalex.org/W7135202318","https://openalex.org/W7140174433","https://openalex.org/W7165031371","https://openalex.org/W7170033841","https://openalex.org/W7171958446","https://openalex.org/W7201877406"],"related_works":["https://openalex.org/W3215421476","https://openalex.org/W4414105366","https://openalex.org/W4415623275","https://openalex.org/W7053289017","https://openalex.org/W4382680941","https://openalex.org/W4413146471","https://openalex.org/W7025312351","https://openalex.org/W4321231430","https://openalex.org/W4319451770","https://openalex.org/W4391306353"],"abstract_inverted_index":{"Deterministic":[0],"neural":[1,4,119],"networks":[2,54],"and":[3,72,90,113,159,204,218],"operators":[5],"provide":[6],"point":[7],"predictions":[8,203,221],"with":[9,184,195],"no":[10],"intrinsic":[11],"measure":[12],"of":[13,28,93,188,242],"reliability.":[14],"Yet,":[15],"predictive":[16,44,88,118,125,149,180,234],"uncertainty":[17,235],"may":[18],"stem":[19],"from":[20,99],"irreducible":[21],"outcome":[22],"variability,":[23],"finite":[24],"data,":[25],"or":[26,55,167,240],"limitations":[27],"the":[29,87,91,117,129,140,155,185],"chosen":[30],"model":[31,61,156,169],"class.":[32],"Monte":[33,196],"Carlo":[34,197],"dropout":[35],"offers":[36,228],"a":[37,43,58,70,100,110,123,135,174,229],"computationally":[38],"convenient":[39],"way":[40,230],"to":[41,223,231],"construct":[42],"distribution":[45],"through":[46,109,116],"stochastic":[47],"feature":[48],"masking,":[49],"without":[50,236],"training":[51,165],"multiple":[52],"independent":[53,102],"explicitly":[56],"inferring":[57],"posterior":[59,238],"over":[60],"parameters.":[62],"However,":[63],"its":[64,94],"perturbation":[65],"law":[66,92],"is":[67,157],"largely":[68],"prescribed":[69],"priori":[71],"typically":[73],"factorised":[74],"across":[75],"latent":[76,95,107],"coordinates.":[77],"We":[78],"introduce":[79],"Optimal":[80],"Transport":[81],"Dropout":[82],"(OTD),":[83],"which":[84],"instead":[85],"learns":[86],"mapping":[89],"perturbations":[96,108],"jointly.":[97],"Starting":[98],"simple":[101],"reference":[103],"distribution,":[104],"OTD":[105,146,199,226],"transports":[106],"learnable":[111],"flow":[112],"propagates":[114],"them":[115],"network,":[120],"thereby":[121],"inducing":[122],"structured":[124,233],"law.":[126],"Training":[127],"uses":[128],"strictly":[130],"proper":[131],"Energy":[132],"Score,":[133],"while":[134],"kinetic-action":[136],"term":[137],"geometrically":[138],"regularises":[139],"transport.":[141],"Synthetic":[142],"benchmarks":[143],"show":[144],"that":[145],"captures":[147],"multimodal":[148],"distributions,":[150],"generates":[151],"meaningful":[152],"dispersion":[153,162,181],"when":[154],"misspecified,":[158],"exhibits":[160],"contracting":[161],"as":[163],"more":[164,201],"data":[166],"greater":[168],"capacity":[170],"are":[171],"provided.":[172],"For":[173],"field-valued":[175],"partial":[176],"differential":[177],"equation":[178],"surrogate,":[179],"strongly":[182],"aligns":[183],"spatial":[186],"pattern":[187],"prediction":[189],"errors.":[190],"On":[191,209],"this":[192],"task,":[193],"compared":[194,222],"dropout,":[198],"yields":[200],"accurate":[202],"better-calibrated,":[205],"substantially":[206],"narrower":[207],"intervals.":[208],"real-world":[210],"regression":[211],"benchmarks,":[212],"it":[213],"further":[214],"shows":[215],"competitive":[216],"accuracy":[217],"better":[219],"probabilistic":[220],"established":[224],"baselines.":[225],"therefore":[227],"learn":[232],"explicit":[237],"inference":[239],"ensembles":[241],"independently":[243],"trained":[244],"predictors.":[245]},"counts_by_year":[],"updated_date":"2026-11-07T11:25:07.637180","created_date":"2026-09-30T00:00:00"}