{"id":"https://openalex.org/W7163133425","doi":"https://doi.org/10.48550/arxiv.2606.00643","title":"Taming the Loss Landscape of PINNs with Noisy Feynman-Kac Supervision: Operator Preconditioning and Non-Asymptotic Error Bounds","display_name":"Taming the Loss Landscape of PINNs with Noisy Feynman-Kac Supervision: Operator Preconditioning and Non-Asymptotic Error Bounds","publication_year":2026,"publication_date":"2026-05-30","ids":{"openalex":"https://openalex.org/W7163133425","doi":"https://doi.org/10.48550/arxiv.2606.00643"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2606.00643","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2606.00643","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":null,"license_id":null,"version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Preprint"},"type":"preprint","indexed_in":["datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://doi.org/10.48550/arxiv.2606.00643","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5117422752","display_name":"Nathanael Tepakbong","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Tepakbong, Nathanael","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5113290951","display_name":"Hanyu Hu","orcid":"https://orcid.org/0000-0002-7239-047X"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Hu, Hanyu","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5137696112","display_name":"Chengyu Liu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Liu, Chengyu","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5137684371","display_name":"Xiang Zhou","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhou, Xiang","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":0,"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/T11206","display_name":"Model Reduction and Neural Networks","score":0.8571000099182129,"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"}},"topics":[{"id":"https://openalex.org/T11206","display_name":"Model Reduction and Neural Networks","score":0.8571000099182129,"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"}},{"id":"https://openalex.org/T11612","display_name":"Stochastic Gradient Optimization Techniques","score":0.04699999839067459,"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":0.01590000092983246,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/pointwise","display_name":"Pointwise","score":0.8704000115394592},{"id":"https://openalex.org/keywords/boundary","display_name":"Boundary (topology)","score":0.48399999737739563},{"id":"https://openalex.org/keywords/artificial-neural-network","display_name":"Artificial neural network","score":0.4616999924182892},{"id":"https://openalex.org/keywords/term","display_name":"Term (time)","score":0.45829999446868896},{"id":"https://openalex.org/keywords/domain","display_name":"Domain (mathematical analysis)","score":0.43650001287460327},{"id":"https://openalex.org/keywords/operator","display_name":"Operator (biology)","score":0.4196000099182129},{"id":"https://openalex.org/keywords/residual","display_name":"Residual","score":0.39820000529289246}],"concepts":[{"id":"https://openalex.org/C2777984123","wikidata":"https://www.wikidata.org/wiki/Q9248237","display_name":"Pointwise","level":2,"score":0.8704000115394592},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.5760999917984009},{"id":"https://openalex.org/C62354387","wikidata":"https://www.wikidata.org/wiki/Q875399","display_name":"Boundary (topology)","level":2,"score":0.48399999737739563},{"id":"https://openalex.org/C28826006","wikidata":"https://www.wikidata.org/wiki/Q33521","display_name":"Applied mathematics","level":1,"score":0.4722000062465668},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.4616999924182892},{"id":"https://openalex.org/C61797465","wikidata":"https://www.wikidata.org/wiki/Q1188986","display_name":"Term (time)","level":2,"score":0.45829999446868896},{"id":"https://openalex.org/C36503486","wikidata":"https://www.wikidata.org/wiki/Q11235244","display_name":"Domain (mathematical analysis)","level":2,"score":0.43650001287460327},{"id":"https://openalex.org/C17020691","wikidata":"https://www.wikidata.org/wiki/Q139677","display_name":"Operator (biology)","level":5,"score":0.4196000099182129},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.4056999981403351},{"id":"https://openalex.org/C155512373","wikidata":"https://www.wikidata.org/wiki/Q287450","display_name":"Residual","level":2,"score":0.39820000529289246},{"id":"https://openalex.org/C19499675","wikidata":"https://www.wikidata.org/wiki/Q232207","display_name":"Monte Carlo method","level":2,"score":0.3569999933242798},{"id":"https://openalex.org/C126255220","wikidata":"https://www.wikidata.org/wiki/Q141495","display_name":"Mathematical optimization","level":1,"score":0.3549000024795532},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.3418999910354614},{"id":"https://openalex.org/C93226319","wikidata":"https://www.wikidata.org/wiki/Q193137","display_name":"Differential (mechanical device)","level":2,"score":0.33489999175071716},{"id":"https://openalex.org/C93779851","wikidata":"https://www.wikidata.org/wiki/Q271977","display_name":"Partial differential equation","level":2,"score":0.31459999084472656},{"id":"https://openalex.org/C2777212361","wikidata":"https://www.wikidata.org/wiki/Q5127848","display_name":"Class (philosophy)","level":2,"score":0.31139999628067017},{"id":"https://openalex.org/C48753275","wikidata":"https://www.wikidata.org/wiki/Q11216","display_name":"Numerical analysis","level":2,"score":0.2671000063419342},{"id":"https://openalex.org/C39891107","wikidata":"https://www.wikidata.org/wiki/Q5767098","display_name":"Hinge loss","level":3,"score":0.25949999690055847},{"id":"https://openalex.org/C70915906","wikidata":"https://www.wikidata.org/wiki/Q1058681","display_name":"Differential operator","level":2,"score":0.2554999887943268}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2606.00643","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2606.00643","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":null,"license_id":null,"version":null,"is_accepted":false,"is_published":null,"raw_source_name":null,"raw_type":"Preprint"}],"best_oa_location":{"id":"doi:10.48550/arxiv.2606.00643","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2606.00643","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":null,"license_id":null,"version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Preprint"},"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":{"Physics-Informed":[0],"Neural":[1],"Networks":[2],"(PINNs)":[3],"often":[4],"train":[5],"slowly":[6],"or":[7],"fail":[8],"to":[9,21,49,169],"converge":[10],"on":[11,178],"challenging":[12],"partial":[13],"differential":[14,30],"equations":[15],"(PDEs),":[16],"a":[17,37,43,74,94,100],"behavior":[18],"recently":[19],"linked":[20],"severely":[22],"ill-conditioned":[23],"loss":[24],"landscapes":[25],"inherited":[26],"from":[27],"the":[28,47,50,81,88,113,124,148,170,188],"underlying":[29],"operator.":[31],"We":[32,56],"study":[33],"PINNs":[34,201],"augmented":[35],"with":[36,136],"pointwise":[38,89],"data-fidelity":[39],"term,":[40],"added":[41],"at":[42],"few":[44],"points":[45],"in":[46,117],"domain":[48],"standard":[51,82,200],"residual":[52],"and":[53,122,156,184,190],"boundary":[54],"losses.":[55],"show":[57,191],"that":[58,192],"this":[59],"supervision":[60],"term":[61],"acts":[62],"as":[63],"an":[64],"operator-level":[65],"preconditioner:":[66],"for":[67,134,154,198],"suitable":[68],"weights,":[69],"our":[70,173],"comparison":[71],"bounds":[72,133,153],"guarantee":[73],"substantially":[75],"smaller":[76],"condition":[77],"number":[78],"than":[79],"under":[80],"PINN":[83],"loss,":[84],"independently":[85],"of":[86,97,112,144,159,165,172],"how":[87],"labels":[90,107],"are":[91,164],"obtained.":[92],"For":[93],"broad":[95],"class":[96],"PDEs":[98,197],"admitting":[99],"Feynman-Kac":[101],"(FK)":[102],"representation,":[103],"we":[104,119,129,150],"generate":[105],"such":[106],"by":[108,140],"Monte":[109],"Carlo":[110],"averages":[111],"FK":[114],"functional,":[115],"resulting":[116],"what":[118],"call":[120],"``FK-PINNs\",":[121],"using":[123],"excess":[125],"risk":[126],"decomposition":[127],"approach,":[128],"derive":[130],"non-asymptotic":[131],"$L^2(\u03a9)$-error":[132],"FK-PINNs":[135,193],"$\\tanh$":[137,160],"activation":[138],"trained":[139],"finitely":[141],"many":[142],"steps":[143],"gradient":[145],"descent.":[146],"Along":[147],"way,":[149],"establish":[151],"pseudo-dimension":[152],"first-":[155],"second-order":[157],"derivatives":[158],"neural":[161],"networks,":[162],"which":[163,199],"independent":[166],"interest":[167],"and,":[168],"best":[171],"knowledge,":[174],"new.":[175],"Numerical":[176],"experiments":[177],"Poisson,":[179],"Schr\u00f6dinger,":[180],"mean":[181],"exit":[182],"time,":[183],"committor":[185],"problems":[186],"corroborate":[187],"theory,":[189],"can":[194],"successfully":[195],"solve":[196],"exhibit":[202],"severe":[203],"failure":[204],"modes.":[205]},"counts_by_year":[],"updated_date":"2026-07-01T06:00:48.157686","created_date":"2026-06-03T00:00:00"}