{"id":"https://openalex.org/W4414360425","doi":"https://doi.org/10.24963/ijcai.2025/947","title":"Robust Finite-Memory Policy Gradients for Hidden-Model POMDPs","display_name":"Robust Finite-Memory Policy Gradients for Hidden-Model POMDPs","publication_year":2025,"publication_date":"2025-09-01","ids":{"openalex":"https://openalex.org/W4414360425","doi":"https://doi.org/10.24963/ijcai.2025/947"},"language":"en","primary_location":{"id":"doi:10.24963/ijcai.2025/947","is_oa":false,"landing_page_url":"https://doi.org/10.24963/ijcai.2025/947","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the Thirty-Fourth International Joint Conference on Artificial Intelligence","raw_type":"proceedings-article"},"type":"article","indexed_in":["crossref"],"open_access":{"is_oa":false,"oa_status":"closed","oa_url":null,"any_repository_has_fulltext":false},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5093536467","display_name":"Maris F. L. Galesloot","orcid":null},"institutions":[{"id":"https://openalex.org/I145872427","display_name":"Radboud University Nijmegen","ror":"https://ror.org/016xsfp80","country_code":"NL","type":"education","lineage":["https://openalex.org/I145872427"]}],"countries":["NL"],"is_corresponding":true,"raw_author_name":"Maris F. L. Galesloot","raw_affiliation_strings":["Radboud University Nijmegen"],"affiliations":[{"raw_affiliation_string":"Radboud University Nijmegen","institution_ids":["https://openalex.org/I145872427"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5047861254","display_name":"Roman Andriushchenko","orcid":"https://orcid.org/0000-0002-1286-934X"},"institutions":[{"id":"https://openalex.org/I60587646","display_name":"Brno University of Technology","ror":"https://ror.org/03613d656","country_code":"CZ","type":"education","lineage":["https://openalex.org/I60587646"]}],"countries":["CZ"],"is_corresponding":false,"raw_author_name":"Roman Andriushchenko","raw_affiliation_strings":["Brno University of Technology"],"affiliations":[{"raw_affiliation_string":"Brno University of Technology","institution_ids":["https://openalex.org/I60587646"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5001379903","display_name":"Milan \u010ce\u0161ka","orcid":"https://orcid.org/0000-0002-0300-9727"},"institutions":[{"id":"https://openalex.org/I60587646","display_name":"Brno University of Technology","ror":"https://ror.org/03613d656","country_code":"CZ","type":"education","lineage":["https://openalex.org/I60587646"]}],"countries":["CZ"],"is_corresponding":false,"raw_author_name":"Milan Ceska","raw_affiliation_strings":["Brno University of Technology"],"affiliations":[{"raw_affiliation_string":"Brno University of Technology","institution_ids":["https://openalex.org/I60587646"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5018941708","display_name":"Sebastian Junges","orcid":"https://orcid.org/0000-0003-0978-8466"},"institutions":[{"id":"https://openalex.org/I145872427","display_name":"Radboud University Nijmegen","ror":"https://ror.org/016xsfp80","country_code":"NL","type":"education","lineage":["https://openalex.org/I145872427"]}],"countries":["NL"],"is_corresponding":false,"raw_author_name":"Sebastian Junges","raw_affiliation_strings":["Radboud University Nijmegen"],"affiliations":[{"raw_affiliation_string":"Radboud University Nijmegen","institution_ids":["https://openalex.org/I145872427"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5012669709","display_name":"Nils Jansen","orcid":"https://orcid.org/0000-0003-1318-8973"},"institutions":[{"id":"https://openalex.org/I904495901","display_name":"Ruhr University Bochum","ror":"https://ror.org/04tsk2644","country_code":"DE","type":"education","lineage":["https://openalex.org/I904495901"]},{"id":"https://openalex.org/I145872427","display_name":"Radboud University Nijmegen","ror":"https://ror.org/016xsfp80","country_code":"NL","type":"education","lineage":["https://openalex.org/I145872427"]}],"countries":["DE","NL"],"is_corresponding":false,"raw_author_name":"Nils Jansen","raw_affiliation_strings":["Radboud University Nijmegen","Ruhr-University Bochum"],"affiliations":[{"raw_affiliation_string":"Radboud University Nijmegen","institution_ids":["https://openalex.org/I145872427"]},{"raw_affiliation_string":"Ruhr-University Bochum","institution_ids":["https://openalex.org/I904495901"]}]}],"institutions":[],"countries_distinct_count":3,"institutions_distinct_count":5,"corresponding_author_ids":["https://openalex.org/A5093536467"],"corresponding_institution_ids":["https://openalex.org/I145872427"],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.1399944,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"8518","last_page":"8526"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11612","display_name":"Stochastic Gradient Optimization Techniques","score":0.9531000256538391,"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/T11612","display_name":"Stochastic Gradient Optimization Techniques","score":0.9531000256538391,"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/T11321","display_name":"Error Correcting Code Techniques","score":0.9437000155448914,"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/T12288","display_name":"Optimization and Search Problems","score":0.9430000185966492,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/partially-observable-markov-decision-process","display_name":"Partially observable Markov decision process","score":0.7024999856948853},{"id":"https://openalex.org/keywords/markov-decision-process","display_name":"Markov decision process","score":0.6326000094413757},{"id":"https://openalex.org/keywords/subgradient-method","display_name":"Subgradient method","score":0.6079000234603882},{"id":"https://openalex.org/keywords/intuition","display_name":"Intuition","score":0.5680000185966492},{"id":"https://openalex.org/keywords/observable","display_name":"Observable","score":0.5120000243186951},{"id":"https://openalex.org/keywords/set","display_name":"Set (abstract data type)","score":0.5073999762535095},{"id":"https://openalex.org/keywords/robustness","display_name":"Robustness (evolution)","score":0.4984999895095825}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7228999733924866},{"id":"https://openalex.org/C17098449","wikidata":"https://www.wikidata.org/wiki/Q176814","display_name":"Partially observable Markov decision process","level":4,"score":0.7024999856948853},{"id":"https://openalex.org/C106189395","wikidata":"https://www.wikidata.org/wiki/Q176789","display_name":"Markov decision process","level":3,"score":0.6326000094413757},{"id":"https://openalex.org/C158968445","wikidata":"https://www.wikidata.org/wiki/Q7631150","display_name":"Subgradient method","level":2,"score":0.6079000234603882},{"id":"https://openalex.org/C132010649","wikidata":"https://www.wikidata.org/wiki/Q189222","display_name":"Intuition","level":2,"score":0.5680000185966492},{"id":"https://openalex.org/C126255220","wikidata":"https://www.wikidata.org/wiki/Q141495","display_name":"Mathematical optimization","level":1,"score":0.5224000215530396},{"id":"https://openalex.org/C32848918","wikidata":"https://www.wikidata.org/wiki/Q845789","display_name":"Observable","level":2,"score":0.5120000243186951},{"id":"https://openalex.org/C177264268","wikidata":"https://www.wikidata.org/wiki/Q1514741","display_name":"Set (abstract data type)","level":2,"score":0.5073999762535095},{"id":"https://openalex.org/C63479239","wikidata":"https://www.wikidata.org/wiki/Q7353546","display_name":"Robustness (evolution)","level":3,"score":0.4984999895095825},{"id":"https://openalex.org/C31531917","wikidata":"https://www.wikidata.org/wiki/Q915157","display_name":"Robust control","level":3,"score":0.4108000099658966},{"id":"https://openalex.org/C2780791683","wikidata":"https://www.wikidata.org/wiki/Q846785","display_name":"Action (physics)","level":2,"score":0.41040000319480896},{"id":"https://openalex.org/C159886148","wikidata":"https://www.wikidata.org/wiki/Q176645","display_name":"Markov process","level":2,"score":0.36090001463890076},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.35420000553131104},{"id":"https://openalex.org/C98763669","wikidata":"https://www.wikidata.org/wiki/Q176645","display_name":"Markov chain","level":2,"score":0.3215000033378601},{"id":"https://openalex.org/C193254401","wikidata":"https://www.wikidata.org/wiki/Q2160088","display_name":"Robust optimization","level":2,"score":0.2816999852657318},{"id":"https://openalex.org/C97541855","wikidata":"https://www.wikidata.org/wiki/Q830687","display_name":"Reinforcement learning","level":2,"score":0.27619999647140503},{"id":"https://openalex.org/C150325174","wikidata":"https://www.wikidata.org/wiki/Q4335500","display_name":"Optimal decision","level":3,"score":0.2603999972343445},{"id":"https://openalex.org/C91575142","wikidata":"https://www.wikidata.org/wiki/Q1971426","display_name":"Optimal control","level":2,"score":0.25189998745918274}],"mesh":[],"locations_count":2,"locations":[{"id":"doi:10.24963/ijcai.2025/947","is_oa":false,"landing_page_url":"https://doi.org/10.24963/ijcai.2025/947","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the Thirty-Fourth International Joint Conference on Artificial Intelligence","raw_type":"proceedings-article"},{"id":"pmh:oai:repository.ubn.ru.nl:2066/324649","is_oa":false,"landing_page_url":"https://hdl.handle.net/2066/324649","pdf_url":null,"source":{"id":"https://openalex.org/S4306401067","display_name":"Radboud Repository (Radboud University)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I145872427","host_organization_name":"Radboud University Nijmegen","host_organization_lineage":["https://openalex.org/I145872427"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"","raw_type":"Article in monograph or in proceedings"}],"best_oa_location":null,"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":{"Partially":[0],"observable":[1],"Markov":[2],"decision":[3],"processes":[4],"(POMDPs)":[5],"model":[6,53,67],"specific":[7],"environments":[8],"in":[9,25],"sequential":[10],"decision-making":[11],"under":[12],"uncertainty.":[13],"Critically,":[14],"optimal":[15],"policies":[16,97,149],"for":[17,79,88,132],"POMDPs":[18,29,39],"may":[19],"not":[20],"be":[21,69],"robust":[22,78,96,112,153],"against":[23],"perturbations":[24],"the":[26,51,70,121,129],"environment.":[27],"Hidden-model":[28],"(HM-POMDPs)":[30],"capture":[31],"sets":[32],"of":[33,59,90,167],"different":[34],"environment":[35,71],"models,":[36,61],"that":[37,50,109,150,165],"is,":[38],"with":[40],"a":[41,57,80,104,117,133,169],"shared":[42],"action":[43],"and":[44,62,123,154,160],"observation":[45],"space.":[46],"The":[47,136],"intuition":[48],"is":[49,54,64,77],"true":[52],"hidden":[55],"among":[56],"set":[58],"potential":[60],"it":[63,84],"unknown":[65],"which":[66],"will":[68],"at":[72],"execution":[73],"time.":[74],"A":[75],"policy":[76,113,131],"given":[81],"HM-POMDP":[82],"if":[83],"achieves":[85],"sufficient":[86],"performance":[87],"each":[89],"its":[91],"POMDPs.":[92],"We":[93],"compute":[94],"such":[95],"by":[98,115],"combining":[99],"two":[100],"orthogonal":[101],"techniques:":[102],"(1)":[103,147],"deductive":[105],"formal":[106],"verification":[107],"technique":[108],"supports":[110],"tractable":[111],"evaluation":[114,138],"computing":[116],"worst-case":[118,134],"POMDP":[119],"within":[120],"HM-POMDP,":[122],"(2)":[124,161],"subgradient":[125],"ascent":[126],"to":[127,142,157,163],"optimize":[128],"candidate":[130],"POMDP.":[135],"empirical":[137],"shows":[139],"that,":[140],"compared":[141],"various":[143],"baselines,":[144],"our":[145],"approach":[146],"produces":[148],"are":[151],"more":[152],"generalize":[155],"better":[156],"unseen":[158],"POMDPs,":[159],"scales":[162],"HM-POMDPs":[164],"consist":[166],"over":[168],"hundred":[170],"thousand":[171],"environments.":[172]},"counts_by_year":[],"updated_date":"2026-03-07T16:01:11.037858","created_date":"2025-10-10T00:00:00"}