{"id":"https://openalex.org/W7131414289","doi":"https://doi.org/10.48550/arxiv.2602.19967","title":"Unlearning Noise in PINNs: A Selective Pruning Framework for PDE Inverse Problems","display_name":"Unlearning Noise in PINNs: A Selective Pruning Framework for PDE Inverse Problems","publication_year":2026,"publication_date":"2026-02-23","ids":{"openalex":"https://openalex.org/W7131414289","doi":"https://doi.org/10.48550/arxiv.2602.19967"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2602.19967","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2602.19967","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.2602.19967","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5126815454","display_name":"Yongsheng Chen","orcid":null},"institutions":[],"countries":[],"is_corresponding":true,"raw_author_name":"Chen, Yongsheng","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5126850342","display_name":"Yong Chen","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Chen, Yong","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5126824904","display_name":"Wei Guo","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Guo, Wei","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5064601117","display_name":"Xinghui Zhong","orcid":"https://orcid.org/0000-0003-3681-9598"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhong, Xinghui","raw_affiliation_strings":[],"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":4,"corresponding_author_ids":["https://openalex.org/A5126815454"],"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":0.9821000099182129,"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":0.9821000099182129,"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/T11948","display_name":"Machine Learning in Materials Science","score":0.0035000001080334187,"subfield":{"id":"https://openalex.org/subfields/2505","display_name":"Materials Chemistry"},"field":{"id":"https://openalex.org/fields/25","display_name":"Materials Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T10775","display_name":"Generative Adversarial Networks and Image Synthesis","score":0.003100000089034438,"subfield":{"id":"https://openalex.org/subfields/1707","display_name":"Computer Vision and Pattern Recognition"},"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/pruning","display_name":"Pruning","score":0.645799994468689},{"id":"https://openalex.org/keywords/measure","display_name":"Measure (data warehouse)","score":0.5333999991416931},{"id":"https://openalex.org/keywords/artificial-neural-network","display_name":"Artificial neural network","score":0.5315999984741211},{"id":"https://openalex.org/keywords/stability","display_name":"Stability (learning theory)","score":0.47269999980926514},{"id":"https://openalex.org/keywords/inverse-problem","display_name":"Inverse problem","score":0.46790000796318054},{"id":"https://openalex.org/keywords/noise-reduction","display_name":"Noise reduction","score":0.4596000015735626},{"id":"https://openalex.org/keywords/noise","display_name":"Noise (video)","score":0.45329999923706055},{"id":"https://openalex.org/keywords/inverse","display_name":"Inverse","score":0.41019999980926514},{"id":"https://openalex.org/keywords/reliability","display_name":"Reliability (semiconductor)","score":0.3546999990940094}],"concepts":[{"id":"https://openalex.org/C108010975","wikidata":"https://www.wikidata.org/wiki/Q500094","display_name":"Pruning","level":2,"score":0.645799994468689},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6176000237464905},{"id":"https://openalex.org/C2780009758","wikidata":"https://www.wikidata.org/wiki/Q6804172","display_name":"Measure (data warehouse)","level":2,"score":0.5333999991416931},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.5315999984741211},{"id":"https://openalex.org/C112972136","wikidata":"https://www.wikidata.org/wiki/Q7595718","display_name":"Stability (learning theory)","level":2,"score":0.47269999980926514},{"id":"https://openalex.org/C135252773","wikidata":"https://www.wikidata.org/wiki/Q1567213","display_name":"Inverse problem","level":2,"score":0.46790000796318054},{"id":"https://openalex.org/C163294075","wikidata":"https://www.wikidata.org/wiki/Q581861","display_name":"Noise reduction","level":2,"score":0.4596000015735626},{"id":"https://openalex.org/C99498987","wikidata":"https://www.wikidata.org/wiki/Q2210247","display_name":"Noise (video)","level":3,"score":0.45329999923706055},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.4481000006198883},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.41609999537467957},{"id":"https://openalex.org/C207467116","wikidata":"https://www.wikidata.org/wiki/Q4385666","display_name":"Inverse","level":2,"score":0.41019999980926514},{"id":"https://openalex.org/C43214815","wikidata":"https://www.wikidata.org/wiki/Q7310987","display_name":"Reliability (semiconductor)","level":3,"score":0.3546999990940094},{"id":"https://openalex.org/C2776459999","wikidata":"https://www.wikidata.org/wiki/Q2119376","display_name":"Fidelity","level":2,"score":0.3531999886035919},{"id":"https://openalex.org/C177264268","wikidata":"https://www.wikidata.org/wiki/Q1514741","display_name":"Set (abstract data type)","level":2,"score":0.35089999437332153},{"id":"https://openalex.org/C111335779","wikidata":"https://www.wikidata.org/wiki/Q3454686","display_name":"Reduction (mathematics)","level":2,"score":0.34689998626708984},{"id":"https://openalex.org/C2781170535","wikidata":"https://www.wikidata.org/wiki/Q30587856","display_name":"Noisy data","level":2,"score":0.3409999907016754},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.3398999869823456},{"id":"https://openalex.org/C2777303404","wikidata":"https://www.wikidata.org/wiki/Q759757","display_name":"Convergence (economics)","level":2,"score":0.3287999927997589},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.32659998536109924},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.30869999527931213},{"id":"https://openalex.org/C159694833","wikidata":"https://www.wikidata.org/wiki/Q2321565","display_name":"Iterative method","level":2,"score":0.30239999294281006},{"id":"https://openalex.org/C93779851","wikidata":"https://www.wikidata.org/wiki/Q271977","display_name":"Partial differential equation","level":2,"score":0.30000001192092896},{"id":"https://openalex.org/C51632099","wikidata":"https://www.wikidata.org/wiki/Q3985153","display_name":"Training set","level":2,"score":0.2888000011444092},{"id":"https://openalex.org/C151201525","wikidata":"https://www.wikidata.org/wiki/Q177239","display_name":"Limit (mathematics)","level":2,"score":0.27619999647140503},{"id":"https://openalex.org/C137836250","wikidata":"https://www.wikidata.org/wiki/Q984063","display_name":"Optimization problem","level":2,"score":0.27070000767707825},{"id":"https://openalex.org/C29265498","wikidata":"https://www.wikidata.org/wiki/Q7047719","display_name":"Noise measurement","level":3,"score":0.2703999876976013},{"id":"https://openalex.org/C126255220","wikidata":"https://www.wikidata.org/wiki/Q141495","display_name":"Mathematical optimization","level":1,"score":0.2703000009059906},{"id":"https://openalex.org/C160920958","wikidata":"https://www.wikidata.org/wiki/Q7662746","display_name":"Synthetic data","level":2,"score":0.2558000087738037},{"id":"https://openalex.org/C177769412","wikidata":"https://www.wikidata.org/wiki/Q278090","display_name":"Prior probability","level":3,"score":0.2542000114917755},{"id":"https://openalex.org/C12725497","wikidata":"https://www.wikidata.org/wiki/Q810247","display_name":"Baseline (sea)","level":2,"score":0.2524000108242035}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2602.19967","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2602.19967","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.2602.19967","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2602.19967","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":{"Physics-informed":[0],"neural":[1,54],"networks":[2],"(PINNs)":[3],"provide":[4],"a":[5,26,45,76,89,95,104,109,131,187,193,219,238],"promising":[6,239],"framework":[7,79],"for":[8,241],"solving":[9],"inverse":[10,36],"problems":[11,37],"governed":[12],"by":[13,18,63,152,167],"partial":[14],"differential":[15],"equations":[16],"(PDEs)":[17],"integrating":[19],"observational":[20],"data":[21,87,113,178],"and":[22,59,69,115,125,210],"physical":[23],"constraints":[24],"in":[25,66,88,222,248],"unified":[27],"optimization":[28],"objective.":[29],"However,":[30],"the":[31,83,99,120,142,176,181,243],"ill-posed":[32],"nature":[33],"of":[34,48,85,112,245],"PDE":[35,116,183,200],"makes":[38],"them":[39],"highly":[40],"sensitive":[41],"to":[42,81,118,180,218],"noise.":[43],"Even":[44],"small":[46],"fraction":[47],"corrupted":[49,86,126,153],"observations":[50],"can":[51],"distort":[52],"internal":[53],"representations,":[55],"severely":[56],"impairing":[57],"accuracy":[58],"destabilizing":[60],"training.":[61],"Motivated":[62],"recent":[64],"advances":[65],"machine":[67],"unlearning":[68],"structured":[70],"network":[71,172],"pruning,":[72],"we":[73,129],"propose":[74],"P-PINN,":[75],"selective":[77],"pruning":[78,160,236],"designed":[80],"unlearn":[82],"influence":[84],"pretrained":[90],"PINN.":[91],"Specifically,":[92],"starting":[93],"from":[94],"PINN":[96],"trained":[97],"on":[98,156,175,198],"full":[100],"dataset,":[101],"P-PINN":[102,205],"evaluates":[103],"joint":[105],"residual--data":[106],"fidelity":[107],"indicator,":[108],"weighted":[110],"combination":[111],"misfit":[114],"residuals,":[117],"partition":[119],"training":[121,211],"set":[122],"into":[123],"reliable":[124,177],"subsets.":[127],"Next,":[128],"introduce":[130],"bias-based":[132],"neuron":[133],"importance":[134],"measure":[135],"that":[136,204,232],"quantifies":[137],"directional":[138],"activation":[139],"discrepancies":[140],"between":[141],"two":[143],"subsets,":[144],"identifying":[145],"neurons":[146,165],"whose":[147],"representations":[148],"are":[149],"predominantly":[150],"driven":[151],"samples.":[154],"Building":[155],"this,":[157],"an":[158],"iterative":[159],"strategy":[161],"then":[162],"removes":[163],"noise-sensitive":[164],"layer":[166],"layer.":[168],"The":[169],"resulting":[170],"pruned":[171],"is":[173,237],"fine-tuned":[174],"subject":[179],"original":[182],"constraints,":[184],"acting":[185],"as":[186],"lightweight":[188],"post-processing":[189],"stage":[190],"rather":[191],"than":[192],"complete":[194],"retraining.":[195],"Numerical":[196],"experiments":[197],"extensive":[199],"inverse-problem":[201],"benchmarks":[202],"demonstrate":[203],"substantially":[206],"improves":[207],"robustness,":[208],"accuracy,":[209],"stability":[212],"under":[213],"noisy":[214],"conditions,":[215],"achieving":[216],"up":[217],"96.6\\%":[220],"reduction":[221],"relative":[223],"error":[224],"compared":[225],"with":[226],"baseline":[227],"PINNs.":[228],"These":[229],"results":[230],"indicate":[231],"activation-level":[233],"post":[234],"hoc":[235],"mechanism":[240],"enhancing":[242],"reliability":[244],"physics-informed":[246],"learning":[247],"noise-contaminated":[249],"settings.":[250]},"counts_by_year":[],"updated_date":"2026-02-26T06:34:08.959763","created_date":"2026-02-26T00:00:00"}